| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Prince Henry’s Institute of Medical Research, Monash Medical Center (H.G.B.), Clayton, Victoria 3168; and Office for Gender and Health (E.C.D., L.D.) and Center for Genetic Epidemiology (J.C., J.L.H.), University of Melbourne, Victoria 3101, Australia
Address all correspondence and requests for reprints to: Prof. H.G. Burger, Prince Henry’s Institute of Medical Research, Level 4, Block E, Monash Medical Center, 246 Clayton Road, Clayton, Victoria 3168, Australia. E-mail: henry.burger{at}med.monash.edu.au
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The Melbourne Women’s Midlife Health Project (9, 10) is a prospective study of biological and lifestyle-related factors associated with the menopause transition in a cohort of middle-aged Australian-born women recruited by random digit telephone dialing from the Melbourne community. It has provided the opportunity to examine longitudinally, levels of total T, sex hormone-binding globulin (SHBG), the calculated free androgen index (FAI), and DHEAS with time relative to final menses, age, and body mass index (BMI).
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Final menstrual period
A woman’s menstrual history during her transition to menopause was documented using menstrual calendars. The calendars were completed between interviews and recorded details of each menstrual period. The day of a woman’s final menstrual period (FMP) was defined retrospectively after 12 consecutive months of amenorrhea.
Hormonal assays
Fasting blood samples were drawn between the fourth and eighth days of the menstrual cycle in regularly cycling women or after 3 months of amenorrhea. SHBG and DHEAS were measured by automated enzyme immunoassay using the Immulite system purchased in kit form from Diagnostic Products (Los Angeles, CA). T and E2 were measured as described previously (10). The FAI was calculated as the ratio of the measured T to measured SHBG x 100. Samples below assay sensitivity were given the value of assay sensitivity: E2, 20 pmol (n = 343; 33%); and T, 0.1 nmol (n = 3,0.3%).
Study sample
The present report is based on those 172 women who had experienced a natural menopause and provided at least 1 blood sample during follow-up and for whom the date of FMP was recorded. Excluded were women who experienced surgical menopause through hysterectomy, bilateral oophorectomy, or endometrial or iatrogenic ablation or who took hormone therapy before the cessation of menses. For women who took hormone therapy after their FMP, all observations taken during hormone therapy were excluded.
Statistical analysis
The women experienced menopause at varying times in the 7 yr of follow-up. By aligning each measurement from a woman according to the date of her FMP, we were able to summarize the accumulated androgen data over a 12-yr time scale (from 5 yr before to 7 yr after FMP).
Modeling of log(SHBG), log(T), log(FAI), and log(DHEAS) as a function of time to menopause
The longitudinal data were analyzed as repeated measures ordered
by time relative to FMP (t), taking into account any correlation
between successive observations within individuals. The hormone values
were transformed by natural logarithm before analysis to ensure that
the distributions of the residuals (observed values - fitted
values) were close to normal. The statistical package FISHER (11) was
used to model the means of the log-transformed hormone levels, µ, as
a function of time, t (measured in years), according to the double
logistic equation: µ(t) = h1 -
2(h1 -
h
)/{exp[
0(t
- )] + exp[1(t - )]} +
ß1age + ß2BMI +
ß2log(E2), where is
the approximate time of maximum rate of change,
h1 is the maximum mean,
h is the mean at approximate time of maximum
change, and 0 and 1
are parameters representing rates of change (12, 13). Similar
methodology was used in a previous analysis of hormone levels (14). The
double logistic model was compared with a straight line regression
model using the Akaike information criteria (15). The regression
coefficients, ß1, ß2,
and ß3, represent the linear effects of age,
BMI, and log(E2), respectively, when adjusting
for time relative to FMP. Results are presented from the best fitting
model only. The correlation between the residual repeated measures
after adjusting the mean for time and possibly age, BMI, and
log(E2) was estimated as a function of the
absolute time between repeated measures, tj -
tk , where tj and
tk represent the times of measures, and is a
measure of the tracking of androgen levels in an individual about the
fitted mean. Tracking refers to the extent to which individuals above
(below) the fitted mean remain above (below) the mean as time
progresses, where a value of 1 means they move parallel to the fitted
mean, and 0 means they vary randomly about the fitted mean.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
During follow-up, a total of 1046 blood samples were taken (average,
6.1/woman). Sixty-four percent of women contributed a complete set of 7
blood samples. Table 1
shows the median
values and normal ranges of the androgen levels, as defined by the 5th
and 95th percentiles, pre- and post-FMP, and for all samples
combined.
|
The double logistic model provided a better fit than the linear
regression model (P < 0.0001), and the estimates are
presented in Table 2. Figure 1A shows the fitted curve against time
relative to FMP. Mean SHBG levels decreased gradually, from 83.8 nmol/L
at -4 yr to 54.5 nmol/L at the FMP and further to 47.7 nmol/L at +2
yr, representing a 43% decrease over 6 yr. The time of maximum change
was estimated to be 2 yr before FMP [95% confidence interval (CI),
0.8–3.2]. We added other variables to the model and found that SHBG
was negatively associated with BMI [ß = -0.039;
SE(ß) = 0.004; P <
0.0001] and positively associated with log(E2)
[ß = 0.069; SE(ß) = 0.010;
P < 0.0001]. After adjusting for these covariates,
the fitted curve (Fig. 1B) appeared flatter; there was only a 29%
decline over 6 yr, and the decline was more abrupt between -4 and -2
yr. About one third of the decline in mean SHBG levels was explained by
adjusting for E2 and BMI. Figure 2 shows the negative linear association
between log(SHBG) and BMI. Correlations between repeated measures of
SHBG were strong within a subject and decreased minimally over time,
from r = 0.7 for measures taken 1 yr apart to r = 0.6 for
measurements taken 6 yr apart (see Fig. 7a).
|
|
|
|
The linear regression model provided the best fit (see Table 3 for the parameter estimates). Mean
levels of T (
1.4 nmol/L) did not vary across the observed 12-yr time
scale [ß = -0.003; SE(ß) = 0.003;
P > 0.05]. The fitted model is presented in Fig. 3. T was also not related to age
[ß = -0.0011; SE(ß) = 0.0021;
P > 0.05], BMI [ß = -0.0022;
SE(ß) = 0.0052; P >
0.05], or log(E2) [ß = 0.0259;
SE(ß) = 0.0721; P >
0.05]. Repeated measures of T were weakly correlated within a subject
(r = 0.25), and this correlation was independent of the time
between measurements (see Fig. 7b
).
|
|
The double logistic model provided a better fit than the linear
regression model (P < 0.0001; see Table 2 for the
corresponding estimates and Fig. 4 for
the fitted curve). Mean levels of the FAI increased with time relative
to FMP by 80% over 6 yr; from 1.5 at -4 yr to 2.7 at +2 yr. The time
of maximum change was estimated to be 2.2 yr before the FMP (95% CI,
1.2–3.2). The FAI was also positively associated with BMI, with an
average increase of 4% for each unit increase in BMI
(P < 0.0001). The FAI was not related to either age or
E2 after allowing for the other two factors. The
correlations between repeated measures of the FAI were moderately
strong within a subject (r = 0.4) and independent of time between
measurements (see Fig. 7c).
|
Linear regression provided the best fit (see Table 3 for the
parameter estimates and Fig. 5, a and b).
Mean levels of DHEAS were not related to the FMP [ß = 0.014;
SE(ß) = 0.019; P > 0.05], but
decreased with age [ß = -0.015;
SE(ß) = 0.005; P < 0.05]
and BMI [ß = -0.038; SE(ß) =
0.019; P < 0.05]. These estimates correspond to a
1.5% decrease for each year of age and a 3.8% decrease for each unit
increase in BMI. Figure 6 shows the
negative linear association between log(DHEAS) and BMI. Repeated
measures of DHEAS within a given subject were highly correlated over
time (r = 0.9 for 1 yr apart; r = 0.8 for 6 yr apart; see
Fig. 7d).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Previous cross-sectional studies suggested that a fall in total T occurs across the menopause. For example, Rozenberg et al. (4) reported T levels in 449 women aged 40 yr and over, without reporting their menopausal status. Mean T levels were significantly higher in the subjects aged 41–50 yr than in those aged 51–60 yr; the actual values derived by inspection of their data indicated a fall from 1.01 nmol/mL to approximately 0.80 nmol/L. Longcope et al. (5) studied 88 women, aged 45–58 yr, who were within 1 month of FMP or up to 76 months later. No change in total T was observed over that period. In their discussion these researchers referred to the fact that the total T levels in their study (mean ± SE, 0.62 ± 0.02 nmol/L) were significantly less than those in a group of normal women sampled on days 5–7 of their cycle (1.01 ± 0.07 nmol/L). The ages of this group of normal women were not specified. Bancroft and Cawood (3) studied 141 women aged 40–60 yr, each sampled on 4 occasions at weekly intervals. The mean total T level in the postmenopausal group (n = 54), 0.71 nmol/L, was significantly lower than that in their cycling group of subjects (n = 49; 0.97 nmol/L). Thus, from cross-sectional studies it has been inferred that the menopause is associated with a fall in total T levels. In our own preliminary analysis of cross-sectional data obtained in the first year of our study, no significant change in total T was observed with changing menopausal status from regularly cycling to postmenopausal (10). One previous prospective longitudinal study (2) reported a 15% fall in total T, comparing mean levels in 20 subjects, a very small sample size. It is of interest that Ushiroyama and Sugimoto (16) observed no change in the ovarian venous concentrations of T when postmenopausal women were compared with premenopausal women.
It is noteworthy that an earlier study had reported a major fall in both total and free T in normal premenopausal women (17). Investigators in that study sampled 33 healthy, regularly cycling, nonobese women, aged 21–51 yr, and calculated that the expected T concentration in a 40-yr-old woman would be 0.61 nmol/L, approximately 50% that in a 21-yr-old woman (1.3 nmol/L). Failure to account for this decline in testosterone concentrations in regularly cycling women before the menopause would result in inappropriate conclusions about the influence of menopause on T if average levels in older women are compared with those in younger women, particularly those less than 30 yr of age.
It should be noted that very limited previous studies (18) have indicated that a midcycle rise in total T concentrations occurs in regularly cycling women. Mushayandebvu et al. (18) reported that in older reproductive aged women, this midcycle T peak in free concentrations in particular is lost. This observation, however, would not account for our inability to demonstrate any change in T across a 12-yr period encompassing FMP.
The changes in SHBG are also controversial. Some previous studies (2, 19) have demonstrated a clear-cut fall in SHBG concentrations related to the menopause, e.g. a fall of approximately 15% in the data of Rannevik et al. (2). Bancroft and Cawood (3) showed no significant change in SHBG levels in comparing pre-, peri-, and postmenopausal women. Longcope et al. (20), on the other hand, reported an 81% increase in SHBG concentrations in a group of 241 women who were initially pre- or perimenopausal and who were followed for a 3-yr period. Although their estrogen levels declined by 32–45%, there was a substantial increase in SHBG. The present report indicates an overall 43% decrease in SHBG from 4 yr before FMP to 2 yr thereafter. The time of maximum change was estimated to be 2 yr before FMP. The levels of SHBG were negatively associated with BMI and were positively correlated with log(E2). About a third of the decline in mean SHBG levels was explicable by changes in E2 and BMI. There is no obvious explanation for this very substantial difference in results.
The T to SHBG ratio was used to calculate FAI, which has been reported to be a reasonable index of free T concentrations in women (21). In light of the absence of any change in total T and a fall in SHBG, the FAI rose by 80% over 6 yr, from 1.5 at -4 yr to 2.7 at +2 yr. Again, the time of maximum change was approximately 2 yr before FMP. As expected, the FAI was positively associated with BMI, but was not related to age or E2. This increase in FAI needs to be taken into account in any model in which it is proposed that a decline in sexual function associated with the menopause transition or menopause might be etiologically related to changes in free androgen levels.
With regard to DHEAS, a number of studies have demonstrated a progressive decline in its circulating concentrations as a function of age, starting in the mid teens or early twenties (7, 22). The decline is linear and shows no obvious relationship with ovarian function, although a previous study (8) had suggested that ovarian factors not related to E2 might significantly influence this reduction in DHEAS levels independently of age. In the present study it was noted that mean DHEAS levels were not related to FMP, but did decrease significantly with age and with BMI. There was 1.5% decrease for each year of age and a 3.8% decrease for each unit increase in BMI. We saw no relationship with menopausal status.
Several previous studies used small sample sizes and were cross-sectional, or if longitudinal, the use of cross-sectional reduction in the statistical analysis impacted on the results obtained. Thus, this is the first prospective study of a population-based sample through the menopausal transition, which overcomes previous methodological problems of sample derivation, length of follow-up, accurate recording of date of FMP, use of an appropriate statistical technique to capture the repeated measurements, and control for the influence of confounding factors.
In summary, the present study, derived from a large prospective longitudinal cohort, failed to show any change in total T in relation to FMP. A clear-cut fall in SHBG with the associated rise in FAI was observed. There was a fall in DHEAS uninfluenced by the menstrual transition and menopause. It is clear that the interactions between androgens and estrogens across the menopausal transition and postmenopausally deserve further investigation in the pathogenesis of the symptoms related to the menopause and perhaps the pathogenesis of disorders that follow it, such as bone loss.
|
Footnotes |
|---|
Received October 29, 1999.
Revised April 19, 2000.
Accepted May 14, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  I. Janssen, L. H. Powell, S. Crawford, B. Lasley, and K. Sutton-Tyrrell Menopause and the Metabolic Syndrome: The Study of Women's Health Across the Nation Arch Intern Med, July 28, 2008; 168(14): 1568 - 1575. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Davis, M.-A. Papalia, R. J. Norman, S. O'Neill, M. Redelman, M. Williamson, B. G.A. Stuckey, J. Wlodarczyk, K. Gard'ner, and A. Humberstone Safety and Efficacy of a Testosterone Metered-Dose Transdermal Spray for Treating Decreased Sexual Satisfaction in Premenopausal Women: A Randomized Trial Ann Intern Med, April 15, 2008; 148(8): 569 - 577. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Remer Dietary calcium, the 16{alpha}-hydroxyl metabolic pathway of steroids, and sex hormones in blood and urine Am. J. Clinical Nutrition, January 1, 2008; 87(1): 192 - 193. [Full Text] [PDF]
|
|
|
|
 |
|
 J. B. Spencer, M. Klein, A. Kumar, and R. Azziz The Age-Associated Decline of Androgens in Reproductive Age and Menopausal Black and White Women J. Clin. Endocrinol. Metab., December 1, 2007; 92(12): 4730 - 4733. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Mah, D. B. Seligson, A. Li, D. C. Marquez, I. I. Wistuba, Y. Elshimali, M. C. Fishbein, D. Chia, R. J. Pietras, and L. Goodglick Aromatase Expression Predicts Survival in Women with Early-Stage Non Small Cell Lung Cancer Cancer Res., November 1, 2007; 67(21): 10484 - 10490. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Dennerstein, P. Lehert, H. G. Burger, and J. R. Guthrie New findings from non-linear longitudinal modelling of menopausal hormone changes Hum. Reprod. Update, November 1, 2007; 13(6): 551 - 557. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.G. Burger, G.E. Hale, D.M. Robertson, and L. Dennerstein A review of hormonal changes during the menopausal transition: focus on findings from the Melbourne Women's Midlife Health Project Hum. Reprod. Update, November 1, 2007; 13(6): 559 - 565. [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]
|
|
|
|
 |
|
 A. Y. Chang, S. M. Abdullah, T. Jain, H. G. Stanek, S. R. Das, D. K. McGuire, R. J. Auchus, and J. A. de Lemos Associations Among Androgens, Estrogens, and Natriuretic Peptides in Young Women: Observations From the Dallas Heart Study J. Am. Coll. Cardiol., January 2, 2007; 49(1): 109 - 116. [Abstract] [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]
|
|
|
|
 |
|
 J. H. Morrison, R. D. Brinton, P. J. Schmidt, and A. C. Gore Estrogen, Menopause, and the Aging Brain: How Basic Neuroscience Can Inform Hormone Therapy in Women J. Neurosci., October 11, 2006; 26(41): 10332 - 10348. [Full Text] [PDF]
|
|
|
|
 |
|
 M.-J. Chen, W.-S. Yang, J.-H. Yang, C. K. Hsiao, Y.-S. Yang, and H.-N. Ho Low sex hormone-binding globulin is associated with low high-density lipoprotein cholesterol and metabolic syndrome in women with PCOS Hum. Reprod., September 1, 2006; 21(9): 2266 - 2271. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B. Singh, M. L. Lee, I. Sinha-Hikim, M. Kushnir, W. Meikle, A. Rockwood, S. Afework, and S. Bhasin Pharmacokinetics of a Testosterone Gel in Healthy Postmenopausal Women J. Clin. Endocrinol. Metab., January 1, 2006; 91(1): 136 - 144. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Randolph Jr, M. Sowers, I. Bondarenko, E. B. Gold, G. A. Greendale, J. T. Bromberger, S. E. Brockwell, and K. A. Matthews The Relationship of Longitudinal Change in Reproductive Hormones and Vasomotor Symptoms during the Menopausal Transition J. Clin. Endocrinol. Metab., November 1, 2005; 90(11): 6106 - 6112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J Bancroft The endocrinology of sexual arousal J. Endocrinol., September 1, 2005; 186(3): 411 - 427. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Santoro, J. Torrens, S. Crawford, J. E. Allsworth, J. S. Finkelstein, E. B. Gold, S. Korenman, W. L. Lasley, J. L. Luborsky, D. McConnell, et al. Correlates of Circulating Androgens in Mid-Life Women: The Study of Women's Health Across the Nation J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4836 - 4845. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R. Davis, S. L. Davison, S. Donath, and R. J. Bell Circulating Androgen Levels and Self-reported Sexual Function in Women JAMA, July 6, 2005; 294(1): 91 - 96. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Davison, R. Bell, S. Donath, J. G. Montalto, and S. R. Davis Androgen Levels in Adult Females: Changes with Age, Menopause, and Oophorectomy J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 3847 - 3853. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bjornerem, B. Straume, M. Midtby, V. Fonnebo, J. Sundsfjord, J. Svartberg, G. Acharya, P. Oian, and G. K. R. Berntsen Endogenous Sex Hormones in Relation to Age, Sex, Lifestyle Factors, and Chronic Diseases in a General Population: The Tromso Study J. Clin. Endocrinol. Metab., December 1, 2004; 89(12): 6039 - 6047. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L.M. Rivera-Woll, M. Papalia, S.R. Davis, and H.G. Burger Androgen insufficiency in women: diagnostic and therapeutic implications Hum. Reprod. Update, September 1, 2004; 10(5): 421 - 432. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P. Nasrallah and B. M. Arafah The Value of Dehydroepiandrosterone Sulfate Measurements in the Assessment of Adrenal Function J. Clin. Endocrinol. Metab., November 1, 2003; 88(11): 5293 - 5298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Piltonen, R. Koivunen, A. Ruokonen, and J. S. Tapanainen Ovarian Age-Related Responsiveness to Human Chorionic Gonadotropin J. Clin. Endocrinol. Metab., July 1, 2003; 88(7): 3327 - 3332. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Morley and H. M. Perry III Androgens and Women at the Menopause and Beyond J. Gerontol. A Biol. Sci. Med. Sci., May 1, 2003; 58(5): M409 - 416. [Full Text] [PDF]
|
|
|
|
|
|
J. F. Randolph Jr., M. Sowers, E. B. Gold, B. A. Mohr, J. Luborsky, N. Santoro, D. S. McConnell, J. S. Finkelstein, S. G. Korenman, K. A. Matthews, et al. Reproductive Hormones in the Early Menopausal Transition: Relationship to Ethnicity, Body Size, and Menopausal Status J. Clin. Endocrinol. Metab., April 1, 2003; 88(4): 1516 - 1522. [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]
|
|
|
|
 |
|
 J. Pfeilschifter, R. Koditz, M. Pfohl, and H. Schatz Changes in Proinflammatory Cytokine Activity after Menopause Endocr. Rev., February 1, 2002; 23(1): 90 - 119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. G. Burger, E. C. Dudley, D. M. Robertson, and L. Dennerstein Hormonal Changes in the Menopause Transition Recent Prog. Horm. Res., January 1, 2002; 57(1): 257 - 275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Couzinet, G. Meduri, M. G. Lecce, J. Young, S. Brailly, H. Loosfelt, E. Milgrom, and G. Schaison The Postmenopausal Ovary Is Not a Major Androgen-Producing Gland J. Clin. Endocrinol. Metab., October 1, 2001; 86(10): 5060 - 5066. [Abstract] [Full Text] [PDF]
|
|
| ||||||||
