| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Obstetrics, Gynecology and Women’s Health (N.S.), Albert Einstein College of Medicine, Bronx, New York 10461; Massachusetts General Hospital (J.S.F.), Boston, Massachusetts 02114; Rush University, Rush-Presbyterian-St. Luke’s Medical Center (J.L.L.), Chicago, Illinois 60612; University of California, Davis/Kaiser (E.B.G., W.L.L.), Davis, California 95616; University of California, Los Angeles (S.K.), Los Angeles, California 90095; University of Medicine and Dentistry/New Jersey Medical School (N.S., J.T., G.W.), Newark, New Jersey 07103; Laboratory: University of Michigan (D.M., M.F.S.), Ann Arbor, Michigan 48104; and New England Research Institutes (S.C., J.E.A.), Watertown, Massachusetts 02472
Address all correspondence and requests for reprints to: Nanette Santoro, M.D., Division of Reproductive Endocrinology, Department of Obstetrics, Gynecology and Women’s Health, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Mazer 316, Bronx, New York 10461. E-mail: glicktoro{at}aol.com.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The aim of this study was to assess the association between androgens and a variety of end points thought to be affected by androgens.
Design: In a community-based baseline cohort of women aged 42–52 yr from the Study of Women’s Health Across the Nation, we measured circulating testosterone (T), dehydroepiandrosterone sulfate, and SHBG, and calculated a free androgen index (FAI) in 2961 women.
Main Outcome Measures: Correlations of androgen measures with each other and with body mass index, waist circumference, and waist-hip ratio were computed, and odds ratios (OR) were estimated for the categorical outcomes of functional limitations, functional status, self-reported health, scores indicative of depressed mood, quality of life, sexual desire and arousal, and the presence of the metabolic syndrome.
Results: Androgens, and particularly SHBG, were associated most strongly with body mass index, waist circumference, and waist-hip ratio. SHBG was associated prominently inversely with the metabolic syndrome (OR = 0.32; 95% confidence interval = 0.26–0.39), which was present in 17% of women at baseline. Dehydroepiandrosterone sulfate was associated modestly with functional status and self-reported health. T was associated minimally with increased sexual desire (OR = 1.09; 95% confidence interval = 1.00–1.18). The association of FAI with self-reported health and depressive symptomatology based on the Center for Epidemiologic Studies Depression Scale score was explained more by T than by SHBG, whereas the association of FAI with sexual arousal and metabolic syndrome was due more to SHBG than to T.
Conclusions: Circulating SHBG and androgens are most strongly associated with physical characteristics and the metabolic syndrome in women in this community-based cohort. Androgens are related weakly to physical functioning and other symptoms to which they commonly are attributed, such as sexual desire, sexual arousal, and well-being.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To determine whether the relative decline in androgens of both adrenal and ovarian origin is associated with adverse symptoms in women, and whether circulating androgens and their determinants are linked to physical characteristics, physical functioning, mood, and sexual interest in women, we analyzed the relationship between serum T, the FAI, SHBG, and dehydroepiandrosterone sulfate (DHEAS), and these factors at baseline in a cohort of 2961 women from a multiethnic, multisite U.S. study—the Study of Women’s Health Across the Nation (SWAN). We hypothesized that women with the highest circulating androgen concentrations would be most likely to manifest the metabolic syndrome, a constellation of obesity-associated insulin resistance and cardiovascular risk states, including dyslipidemia. Secondarily, we hypothesized that women with the lowest circulating androgen concentrations would report a lower sense of well being, more perceived stress, poorer physical functioning, and lower libido than women with higher androgen levels.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SWAN is a multisite, multiethnic longitudinal study of mid-life women (6). Among the goals of SWAN is the characterization of reproductive hormone patterns as women approach and traverse menopause. Women enrolled in the SWAN longitudinal study were recruited from community-based samples drawn from a larger, cross sectional sample of women ages 40–55 yr in geographic areas affiliated with the clinical sites located in seven communities or cities nationwide. At each of the sites, a Caucasian sample and a prespecified non-Caucasian sample were recruited. African-American women were recruited at the Detroit area, Boston, Chicago, and Pittsburgh sites. Hispanic women were recruited at the Newark site. Chinese-American and Japanese American women were recruited at the Oakland, California area and Los Angeles sites, respectively.
SWAN longitudinal study participants were required to: 1) be aged 42–52 yr at baseline; 2) have an intact uterus and at least one ovary; 3) have had at least one menstrual period within 3 months before recruitment; and 4) not be taking any reproductive hormones for at least 3 months before enrollment. The study was approved by the Institutional Review Board of each of the seven SWAN clinical centers, and all women gave written informed consent.
Menopausal status
The menopausal status of the participants was determined at baseline based upon the perceived regularity of cycles and length of amenorrhea. Women with menses in the past 3 months without a noted change in pattern in the last 12 months were considered "premenopausal". Women with self-reported increased irregularity of their cycles in the past year, but fewer than 3 months of amenorrhea, were classified as "early perimenopausal". Similar, but not identical definitions (7) have been used widely in epidemiologic and research studies (8), and have been recommended for menstrual cycle research in perimenopausal women (9).
SWAN participants completed an annual interview that included medical, psychological/cognitive, and symptom data collection. Blood pressure, height, weight, waist circumference, and hip circumference were measured using the same carefully standardized procedures across all seven sites. For these analyses, body mass index (BMI) was categorized as normal (<25 kg/m2), overweight (25–27 kg/m2), or obese (>30 kg/m2).
Demographic variables examined included age, income, education, employment, difficulty paying for basics (food, shelter, and heat), and site. Women were asked about their preferred language for reading and speaking, and this language was used for interviews and questionnaires in the Hispanic, Chinese, and Japanese-American groups (10). Current active and passive smoke exposure was assessed based on American Thoracic Society smoking questions (11) and validated questions on passive exposure, to compute the number of smokers exposed multiplied by hours per week of exposure (12).
Physical, sexual, and health status reporting
Participants were asked to fill out a baseline questionnaire that contained information about health status, psychosocial variables, use of both traditional and alternative medications, and sexual interest and practices. For the purpose of this report, we used answers to specific questions about perceived level of health, degree of functional limitations, overall quality of life, physical activity level compared with their peers, depressive symptomatology, and sexual interest and activity (including desire, arousal, and frequency). All of these measures have been previously validated in large, representative samples of women of a variety of ethnic backgrounds.
Self-reported health was rated on a five-point scale from poor to excellent (13).
Depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale (CES-D). The CES-D is a 20-item depression symptom scale developed by the Center for Epidemiologic Studies (14, 15, 16, 17). A score of 16 or greater was used as the cutoff point to define the presence of depressive symptomatology.
Global quality of life was assessed by the Ladder of Life scale (18), which asks participants to evaluate their current quality of life on a scale of 1–10, with 1 being the worst quality and 10 being the best.
Physical functioning was measured using the SF-36 scale (19, 20, 21, 22). Physical functioning was assessed using an ordinal, three-category variable based on the physical functioning scale, which is a subscale of the larger instrument, the Medical Outcomes Scale (MOS-SF-36). The scale reflects the difficulty of undertaking 10 physical activities that range from vigorous athletic activities to the ability to bathe and dress. An interviewer first asked women whether they were "limited in any way in activities because of any impairments or health problems". If the answer was "no", the MOS-SF-36 physical functioning scale was not administered; otherwise, the physical functioning scale was administered. The responses were scored using norm-based methods and transformed to have a mean of 50 (SD = 100 in the general U.S. population) (1, 19, 23). Because over 85% of the sample had no limitations on physical functioning, we categorized the physical functioning scale as no limitations (maximum score), low limitations (score greater than median of values below the maximum), and high limitations (score less than median of values below the maximum).
Questions on sexual behavior and sexuality were designed to address issues relevant to women in their 40s and 50s. Questions assess sexual behavior and interest for women with and without partners. The questionnaire is derived from items from several sources: The Massachusetts Women’s Health Study (24), the National Health and Social Life Survey (25), the National Survey of Family Growth (26), and the Women’s Health Initiative Daily Life Form (27). These were studies of large representative samples of African American, Asian, Caucasian, and Hispanic women, aged 15–79. Topics included in the surveys covered a wide variety of sexual practices and preferences including homosexual desire, identity and behavior, sexual satisfaction, and sexual dysfunction.
Sexual interest was queried in two ways. Women were asked about the frequency with which they experienced a desire to engage in sex and were asked to report the closest response on the following five-point scale: 1, not at all; 2, once or twice a month; 3, once a week; 4, more than once a week; and 5, daily or more. The frequency with which women experienced arousal during sexual activity was also recorded by self report on a six-point scale: always, almost always, sometimes, almost never, never, and "no activity within the past 6 months". In multivariate analyses, the "almost never" and "never" categories were combined due to small cell counts, and the "no activity" category was omitted.
Blood was drawn annually to measure serum FSH, estradiol (E2), DHEAS, T, and SHBG concentrations. TSH was measured at the baseline visit. Cardiovascular measures, such as total, high-density lipoprotein (HDL) and low-density lipoprotein cholesterol, Lp(a), triglycerides, fasting glucose and insulin, prothrombin time/partial thromboplastin time, and fibrinogen were obtained. This study will focus on the baseline annual, early follicular phase androgens, T and DHEAS, and the metabolic markers of fasting insulin, glucose, and SHBG.
Each woman was scheduled to have an annual blood draw. When possible, blood was drawn between d 2 and 7 of the follicular phase of the menstrual cycle after a minimum 10-h fast. All samples were maintained at 4 C until separated, frozen at 80 C, and then shipped on dry ice to the central laboratories (Central Ligand Assay Satellite Services, University of Michigan, Ann Arbor, MI, and Medical Research Laboratories, Lexington, KY) for analysis. Throughout the study, Medical Research Laboratories participated in and remained certified by the National Heart Lung and Blood Institute, Centers for Disease Control Part III program.
Assay methods
Single hormone assays were conducted using an ACS-180 automated analyzer (Bayer Diagnostics Corp., Norwood, MA). Replicate samples were measured as part of quality control procedures in the earliest years of SWAN, and the decision to abandon duplicate measurements was made by the study’s Laboratory Oversight Committee.
Serum E2 was measured with a modified, off-line ACS-180 (E2–6) immunoassay. Inter- and intraassay coefficients of variation averaged 10.6 and 6.4%, respectively, over the assay range, and the lower limit of detection was 1 pg/ml.
Serum T concentrations were determined by competitive binding of a dimethyl acridinium ester-labeled T derivative to a rabbit polyclonal anti-T antibody premixed with monoclonal antirabbit IgG antibody immobilized on the solid phase paramagnetic particles. Inter- and intraassay coefficients of variation were 10.5 and 8.5%, respectively, and the lower limit of detection was 2 ng/dl. This assay was validated extensively in 1995 against other standard (at the time), state-of-the-art assay systems and was specifically calibrated to be sensitive to the low levels found in women. The assay for SWAN uses the same antibody and reagents used in the Bayer T assay which has been validated and standardized analytically by gas chromatography-mass spectrometry.
The newly established, two-site chemiluminescent assays for serum SHBG and DHEAS concentrations involved competitive binding of dimethyl acridinium ester-labeled SHBG or DHEAS to a commercially available rabbit anti-SHBG or anti-DHEAS antibody and a solid phase of goat antirabbit IgG conjugated to paramagnetic particles. Inter- and intraassay coefficients of variation for the SHBG assay were 9.9 and 6.1%, respectively, and the lower limit of detection was 2 nM. Inter- and intraassay coefficients of variation for the DHEAS assay were 11.3 and 7.6%, respectively, and the lower limit of detection was 2 µg/dl. Total T was indexed to SHBG to calculate the FAI (FAI = 100 x T (ng/dl)/ 28.84 x SHBG (nM).
Insulin, glucose HDL, and low-density lipoprotein cholesterol measurement. All lipid and lipoprotein fractions were analyzed on EDTA-treated plasma. Total cholesterol was analyzed by enzymatic methods. HDL-C was isolated using heparin-2M manganese chloride. Serum insulin was measured using RIA (DPC Coat-a-Count, Los Angeles, CA) procedure and monitored as part of the monthly quality assurance program by the Diabetes Diagnostic Laboratory at the University of Missouri (Columbia, MO). This assay has been shown to have 32% cross-reactivity with proinsulin, and no detectable cross reactivity with C-peptide or glucagon.
Glucose was measured using a hexokinase-coupled reaction (Roche Molecular Biochemicals Diagnostics, Indianapolis, IN).
Metabolic syndrome definition
The National Cholesterol Education Program (NCEP) and the World Health Organization have proposed two distinct definitions of the metabolic syndrome (28, 29). While the NCEP Adult Treatment Panel, Part III definition is most tightly associated with an increased incidence of diabetes mellitus, the World Health Organization definition is more tightly associated with an increased risk of cardiovascular disease when blood pressure is defined as more than 160/90 or more than 140/90, (depending upon the study cited). Within the NCEP criteria, the high normal value of 130–139/85–90 for blood pressure has been associated with an increased risk of heart disease, and this definition is most indicative of the risk for glucose intolerance (30). Because the population to which the NCEP criteria are to be applied (and from which the criteria were developed) contained a large percentage of Asian and Pacific Islanders and were thus similar to the women in SWAN, we classified participants as having the metabolic syndrome if they satisfied three or more of the following criteria: 1) abdominal obesity (waist circumference >80 cm for Chinese and Japanese, >88 cm for Caucasians, African-Americans, and Hispanics); 2) hypertriglyceridemia (fasting triglycerides >150 mg/dl); 3) HDL cholesterol < 50 mg/dl; 4) high blood pressure (systolic blood pressure >130 mm Hg or diastolic blood pressure >85 mm Hg, or on antihypertensive medication); and 5) impaired fasting glucose and or diabetes (fasting glucose >110 mg/dl, or on insulin therapy). Women with the most severe manifestations of glucose intolerance and ß-cell failure, consistent with a diagnosis of diabetes (fasting glucose >126 mg/dl or a history of insulin or oral hypoglycemic agent therapy) were eliminated from this analytic sample, because of potential interactions between a diagnosis of overt diabetes with serum androgens.
Analytic sample
Of the total cohort of 3302 women, 341 were excluded from this analysis: 231 had a fasting blood glucose of at least 126 or reported a diagnosis of diabetes mellitus, 90 had an abnormal screening TSH (<0.1 or >10 µU/ml), 16 did not have data for either T or SHBG, and four were taking exogenous DHEA. Therefore, the final analytic sample included 2961 women, with some additional losses of participants in the multivariate analyses.
Data analysis
Characteristics of the analytic sample were summarized using frequency distributions for categorical variables and medians/interquartile ranges (75th percentile minus 25th percentile) for continuous variables. Pearson correlations between pairs of hormones were computed before and after adjustment for age, site, ethnicity, smoking status (never, past, current), and waist circumference. Pearson correlations between hormones and measures of body composition were computed before and after adjustment for age, site, ethnicity, and smoking status. Associations of hormones with categorical outcomes were estimated in terms of odds ratios (OR) from logistic regression (31), before and after adjustment of age, site ethnicity, smoking status, and waist circumference. For ordinal outcomes not satisfying the assumption of proportional odds, a partial proportional odds logistic regression model was estimated (32). Analyses of metabolic syndrome omitted adjustment for waist circumference, as this is part of the definition. Hormones and body composition measures were log-transformed to handle right skewness. P values of less than 0.05 were considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
shows the baseline characteristics of the SWAN sample in terms of age, ethnicity, physical measurements, functional limitations, self-reported health, CES-D score more than 16, quality of life, smoking status, sexual interest and frequency, and androgens. Nearly 24% of the women (n = 703) had a CES-D score indicative of depressive symptomatology, and 14.7% (425) of the women had a physical functioning score from the SF-36 indicative of physical limitations that interfered with their daily activities. Almost 17% of the overall baseline sample met the definition of the metabolic syndrome. At baseline, the participants were divided almost equally between premenopausal (54.1%) and early perimenopausal (45.9%) status. Most reported that the start of their menstrual periods was predictable to within plus or minus 4 d, i.e. within an 8-d window (79.5%), and 85.6% had a cycle length between 21 and 35 d. Not surprisingly, premenopausal women were highly likely to report predictable periods (92%), although a large proportion of early perimenopausal women also reported predictable periods (71%), suggesting that the degree of irregularity was not yet high in this group.
|
Correlations of log-transformed hormones with each other and with log-transformed measures of body composition are shown in Tables 2 and 3 (unadjusted and adjusted, respectively, for age, ethnicity, site, and smoking status). Log T was correlated moderately positively with log DHEAS, but was not associated with log SHBG. Log DHEAS was correlated weakly negatively with log SHBG, and was associated moderately positively with log FAI. By definition, log FAI was correlated highly positively with log T and highly negatively correlated with log SHBG. Log E2 was related weakly and negatively to BMI, waist-hip ratio (WHR), and waist circumference. Adjustment for log waist circumference, ethnicity, site, age, and smoking had minimal impact on these associations between hormones.
|
|
in association with hormonal measures. There was a strong and consistent negative relationship between log SHBG and increasing BMI. Associations of BMI with log FAI reflected associations with log SHBG (Fig. 1, C and D). Log BMI and log waist circumference were associated weakly positively with log T, and related weakly negatively to log DHEAS (Fig. 1, A and B). The correlation of log WHR and log DHEAS was small and negative, but statistically significant due to the large sample size. These correlations changed only minimally with adjustment for age, ethnicity, site, and smoking.
|
Tables 4 (unadjusted) and 5 (adjusted for log waist circumference, age, ethnicity, site, and smoking status) present the OR for log-transformed hormones from logistic regression models for categorical outcomes, computed as the difference between the 75th and 25th percentiles of each log-transformed hormone. Before adjustment for other factors, T was associated slightly positively with self-reported health [OR = 1.17, 95% confidence interval (95% CI) = 1.08–1.27] and sexual desire (OR = 1.09, 95% CI = 1.00–1.18). Upon adjustment for waist circumference, age, ethnicity, site, and smoking, the relation between T and sexual desire was no longer significant, but T was related significantly negatively to CES-D score. DHEAS was associated positively with both self-reported health and functional status, and related weakly inversely to CES-D scores for depressive symptomatology. The latter association was not significant statistically after adjustment for other factors. FAI was related positively to sexual arousal, reflecting the combination of a positive association of arousal with T and a negative association of arousal with SHBG. After adjustment for other factors, FAI was related significantly positively to self-reported health, sexual desire, and arousal, and related negatively to CES-D score.
|
|
), with an unadjusted OR of 0.32 (95% CI = 0.26–0.39). T was associated modestly with an increased probability of the metabolic syndrome (Fig. 2A) (OR = 1.30, 95% CI = 1.15–1.47). Reflecting these two associations, FAI was related positively to presence of the metabolic syndrome (Fig. 2D) (OR = 2.14, 95% CI = 1.87–2.45). DHEAS was related inversely to the metabolic syndrome (Fig. 2B) OR = 0.82, 95% CI = 0.73–0.92). These relationships persisted in the adjusted analyses. Log E2 was related to the metabolic syndrome in both unadjusted and adjusted analyses (OR = 0.67, 95% CI = 0.59–0.78), but was not related to any of the outcomes related to well-being or libido.
|
Reproductive factors did not bear strong relationships with circulating hormones. The number of live births was slightly likely to be lower in women with higher DHEAS (OR = 0.93, 95% CI = 0.84–0.98) and higher FAI (OR = 0.92 95% CI = 0.84–1.00; data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Androgen bioavailability in women is a function of circulating ovarian and adrenal steroids and their interconversion in target tissues. Bioavailability of circulating androgens is influenced most extensively by the concentrations of the carrier protein, SHBG. SHBG is best known as a carrier protein for circulating E2 and T. It is produced by the liver, and its production is under the physiological control of E2, which positively regulates SHBG, and T and other androgens, which negatively regulate SHBG (33). SHBG is strongly and directly associated with insulin sensitivity: i.e. the greater insulin sensitivity, the higher the circulating SHBG. Because SHBG binds both E2 and T, higher SHBG levels are generally associated with lower levels of the free, or bioactive, fractions of sex steroids. Thus, it is not surprising that the relationships between SHBG and FAI and various outcomes tended to be of similar strength, but in opposite directions. Whether or not SHBG itself is capable of transducing intracellular signals in women is unclear, but a receptor for SHBG has been identified in the prostate (34). The strength of the relationship we observed between SHBG and a "metabolic syndrome" phenotype is likely to represent a combination of the two chief factors that influence circulating SHBG: reproductive hormones (androgens and estrogens) and metabolic factors (insulin sensitivity). In our study, lower SHBG was associated with increased insulin resistance (35) and a greater likelihood of a metabolic syndrome phenotype, as well as increased bioavailable androgens.
A limitation of our study design included the use of a direct T assay. We adjusted for SHBG concentration to calculate a FAI, but did not measure "free" T in serum. While this assay has been modified to render it sufficiently sensitive to the low levels found in women (36), it is possible that extremely low concentrations or extreme mismatches of carrier protein and hormone resulted in over- or underestimation of the true free T level.
Our findings are in general agreement with other studies of middle-aged women. In the Melbourne Women’s Health Project (4), SHBG and DHEAS were associated negatively with BMI, and FAI was associated positively with BMI at a very similar magnitude for all associations. In 102 Caucasian women from the Melbourne Women’s Health Project (37), abdominal adiposity, as measured by dual energy x-ray absorptiometry, was associated with the free T index (FTI). Because in the longitudinal analysis of the Melbourne androgen data there was no relationship between total T and BMI (4), the association of abdominal adiposity with FTI may be best explained as related to insulin resistance and low SHBG. Guthrie et al. (38) in the overall Melbourne sample of 384 women, also observed a relationship between coronary disease risk score and high baseline BMI, as well as increasing BMI, declining E2, and free T. In our study, SHBG, and not androgens, was most strongly related to the presence of the metabolic syndrome in this population of largely normoandrogenic women at baseline. It is possible that aging and progression through the menopausal transition modifies the effects of hormones on the expression of the metabolic syndrome, and consequently upon coronary disease risk. Whether large or increasing body size functions as a metabolic "trigger" that initiates changes in fuel economy at mid-life has been proposed, but is at present, unproven (39, 40). A majority of women in the SWAN study are "overweight" or "obese" (median BMI = 28.1 kg/m2 for the overall cohort and 26.2 kg/m2 for our analytic sample) (41).
We observed an inverse relationship between DHEAS and waist circumference that also was significant statistically, but opposite to the relationship observed with T. These data suggest that androgens of adrenal origin relate to body morphometry in a manner that is opposite to androgens of primarily ovarian origin. Burger et al. (4) have previously reported a positive relationship between FTI and BMI, but a negative relationship between DHEAS and BMI in a smaller, longitudinal sample of perimenopausal women.
Self-reported health was related significantly to both DHEAS and T, and both relationships remained significant after adjustment for log waist circumference, ethnicity, site, age, and smoking status. The robustness of these relationships suggests that some of the positive mood and strength effects associated with T replacement therapy in aging hypogonadal men might also accrue to women (42). The fact that both of these hormones influenced self-reported health in the same direction and at approximately the same magnitude also suggests that this is a direct androgen effect. In some clinical trials (43), DHEA administration improved the sense of well being of peri- and postmenopausal women, although not in others (44). It remains unclear whether endogenous DHEAS levels are simply a surrogate marker of general well being and health, or whether DHEA has a causative effect on well being in an aging population, in whom DHEAS levels decline dramatically (2). Exogenous T improves muscle mass and muscle strength in menopausal women (45). Endogenous T might plausibly improve self-reported health measures if it is related directly to muscle strength and endurance.
We were surprised by the weakness, or even absence, of an association between circulating androgens and physical functional limitations, mood, or sexual function. There are virtually no studies that have examined androgens and self-reported physical functioning among women at the mid-life, although functioning status has been related to other factors associated with circulating androgen, including BMI, WHR, and metabolic syndrome. The relation of physical functioning and androgens is also more complex because the androgen, body size, metabolic syndrome triad is differentially distributed among the race/ethnic groups reported in this study.
There are few epidemiological studies that have examined the relationship of reproductive hormones to sexual interest across the menopausal transition. Our baseline data provide a cross sectional view of pre- and early perimenopausal women. A cross sectional study of 141 women aged 40–60, that used a different method of measurement for sexual function, did not observe a relationship between any of the hormonal variables and sexuality (46). In this latter study, the relatively small sample size and the large age range, spanning the pre- to postmenopausal years, may have obscured some associations. It is possible that the relationships between hormones and sexuality changes over time. Dennerstein et al. (47) have reported a significant association between decreasing circulating E2, but not T or the FAI, with sexual difficulty in an Australian population-based cohort of women who traversed menopause. Another, smaller longitudinal study of a convenience sample of 43 women also related declining E2 levels with decreasing sexual function (48). Our participants were studied when they were either before or in the earliest stages of the menopause transition, and thus E2 would not have been expected to be related to sexual function or motivation. In a Swedish study of 83 women aged 40–53 a relationship between sexuality and androstenedione was observed, but no relationship was noted for T or SHBG (49). Our overall null findings were unaffected by the marital or partnered status of the study participants.
Sexual interest has been reported to improve when exogenous androgens, either in the form of transdermal T (50) or DHEA (51) are given to androgen-deficient individuals. A recent study reported that sexual desire improves when postmenopausal women are given exogenous estrogen plus T (52). Goldstat et al. (53) reported that transdermal T improves general well being and libido in premenopausal women. Yet, in this large sample, we noted only a weak association of T with sexual desire and DHEAS with arousal (which was no longer present after adjustment for other participant characteristics). About 80% of our cohort reported some sexual activity within 6 months of their interview, and were able to provide information about both sexual arousal as well as interest. Therefore, we do not believe that a low level of baseline sexual activity was responsible for this finding. From these results, we cannot conclude that circulating androgens play a significant role in sexual drive in mid-life women undergoing the earliest stages of the menopausal transition. However, it is possible that androgens play a more important role in sexual behavior in postmenopausal women or that they interact with menopausal status as women advance through the transition.
Ovarian androgens fluctuate throughout the menstrual cycle, with a peak at midcycle (5). It is possible that these fluctuations are more important determinants of sexual interest and arousal than an early follicular phase level. Cycle-related changes in androgens would not be addressed by the SWAN blood drawing protocol, which relied upon early follicular phase samples only. It is also possible the 5-
-reduced androgens, produced in target tissues after enzymatic conversion, which are inaccessible for peripheral measurement, are more likely to determine behavioral effects of androgens (54). The tighter linkage of T with its sexual and behavioral correlates in men (55) may be related to the much higher circulating concentrations, which reflect the local "pool" of available hormone for conversion to the more potent 5--reduced androgens more closely.
In summary, we found a modest association between SHBG and increased body size, waist circumference, and WHR, and a strong association of SHBG with the metabolic syndrome in a cross-sectional sample of 2961 women aged 42–52 yr. The relationship between serum T levels and these measures was weaker, and we did not detect consistent relationships between circulating T, FAI, SHBG, and physical functioning, mood, or well being. Self-reported health was related modestly to circulating T and DHEAS levels. Sexual desire was, at best, marginally related to circulating T levels. These data underscore the critical need for improved understanding of the biology of androgens in women, particularly at mid-life and beyond.
|
Acknowledgments |
|---|
Clinical Centers: University of Michigan, Ann Arbor, MI—MaryFran Sowers, principal investigator (PI); Massachusetts General Hospital, Boston, MA—Robert Neer, PI 1995-1999; Joel Finkelstein, PI 1999 to present; Rush University, Rush University Medical Center, Chicago, IL—Lynda Powell, PI; University of California, Davis/Kaiser, Davis, CA—Ellen Gold, PI; University of California, Los Angeles, Los Angeles, CA—Gail Greendale, PI; University of Medicine and Dentistry, New Jersey Medical School, Newark, NJ—Gerson Weiss, PI 1995-2004: Nanette Santoro, PI 2004 to present; and the University of Pittsburgh, Pittsburgh, PA—Karen Matthews, PI.
NIH Program Office: National Institute on Aging, Bethesda, MD—Marcia Ory, 1994-2001; Sherry Sherman, 1994 to present; National Institute of Nursing Research, Bethesda, MD—Program Officers.
Central Laboratory: University of Michigan, Ann Arbor, MI—Daniel McConnell (Central Ligand Assay Satellite Services).
Coordinating Center: New England Research Institutes, Watertown, MA—Sonja McKinlay, PI 1995-2001; University of Pittsburgh, Pittsburgh, PA—Kim Sutton-Tyrrell, PI 2001 to present.
Steering Committee: Chris Gallagher, Chair; Susan Johnson, Chair.
Project Officers: Yvonne Bryan, Taylor Harden, Carole Hudgings, Marcia Ory, and Sheryl Sherman.
We thank the study staff at each site and all the women who participated in SWAN.
Supplemental funding from the National Institute of Mental Health, the National Institute on Child Health and Human Development, the National Center on Complementary and Alternative Medicine, the Office of Minority Health, and the Office of AIDS Research is also gratefully acknowledged. This work was also supported by grants K24 DK02759 from Massachusetts General Hospital (to J.S.F.), K24 HD419782 from Albert Einstein College of Medicine (to N.S.) and K12 HD01457 from UMDNJ–New Jersey Medical School (to J.T.).
|
Footnotes |
|---|
Present address for J.E.A.: Center for Gerontology and Health Care Research, Brown University, Providence, Rhode Island 02912.
This manuscript was reviewed by the Publications and Presentations Committee of The Study of Women’s Health Across the Nation and has its endorsement.
First Published Online April 19, 2005
Abbreviations: BMI, Body mass index; CES-D, Center for Epidemiologic Studies Depression Scale; 95% CI, 95% confidence interval; DHEAS, dehydroepiandrosterone sulfate; E2, estradiol; FAI, free androgen index; FTI, free T index; HDL, high-density lipoprotein; NCEP, National Cholesterol Education Program; OR, odds ratio; SWAN, Study of Women’s Health Across the Nation; T, testosterone; WHR, waist-hip ratio.
Received October 18, 2004.
Accepted April 12, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. Santoro, S. L. Crawford, W. L. Lasley, J. L. Luborsky, K. A. Matthews, D. McConnell, J. F. Randolph Jr., E. B. Gold, G. A. Greendale, S. G. Korenman, et al. Factors Related to Declining Luteal Function in Women during the Menopausal Transition J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1711 - 1721. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Elaut, G. De Cuypere, P. De Sutter, L. Gijs, M. Van Trotsenburg, G. Heylens, J.-M. Kaufman, R. Rubens, and G. T'Sjoen Hypoactive sexual desire in transsexual women: prevalence and association with testosterone levels Eur. J. Endocrinol., March 1, 2008; 158(3): 393 - 399. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. P. Almeida, B. B. Yeap, G. J. Hankey, K. Jamrozik, and L. Flicker Low Free Testosterone Concentration as a Potentially Treatable Cause of Depressive Symptoms in Older Men Arch Gen Psychiatry, March 1, 2008; 65(3): 283 - 289. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. P. Rose, S. M. Haffner, and J. Baillargeon Adiposity, the Metabolic Syndrome, and Breast Cancer in African-American and White American Women Endocr. Rev., December 1, 2007; 28(7): 763 - 777. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. R. Gracia, E. W. Freeman, M. D. Sammel, H. Lin, and M. Mogul Hormones and Sexuality During Transition to Menopause Obstet. Gynecol., April 1, 2007; 109(4): 831 - 840. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Potter A 60-Year-Old Woman With Sexual Difficulties JAMA, February 14, 2007; 297(6): 620 - 633. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Cappola, S. J. Ratcliffe, S. Bhasin, M. R. Blackman, J. Cauley, J. Robbins, J. M. Zmuda, T. Harris, and L. P. Fried Determinants of Serum Total and Free Testosterone Levels in Women over the Age of 65 Years J. Clin. Endocrinol. Metab., February 1, 2007; 92(2): 509 - 516. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Rosner, R. J. Auchus, R. Azziz, P. M. Sluss, and H. Raff Utility, Limitations, and Pitfalls in Measuring Testosterone: An Endocrine Society Position Statement J. Clin. Endocrinol. Metab., February 1, 2007; 92(2): 405 - 413. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Titomanlio, A. Verloes, J.-C. Mercier, A. Schlegel, A. J. Polotsky, J. Rieder, N. Santoro, A. G. Hoppin, and L. M. Kaplan Case 31-2006: A Girl with Severe Obesity N. Engl. J. Med., January 11, 2007; 356(2): 194 - 196. [Full Text] [PDF]
|
|
|
|
|
|
S. Basaria and A. S. Dobs Controversies Regarding Transdermal Androgen Therapy in Postmenopausal Women J. Clin. Endocrinol. Metab., December 1, 2006; 91(12): 4743 - 4752. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Wierman, R. Basson, S. R. Davis, S. Khosla, K. K. Miller, W. Rosner, and N. Santoro Androgen Therapy in Women: An Endocrine Society Clinical Practice Guideline J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3697 - 3710. [Abstract] [Full Text] [PDF]
|
|
