Sex Steroids and Bone: Current Perspectives: Juan Balasch

dmg017
Human Reproduction Update, Vol.9, No.3 pp. 207±222, 2003 DOI: 10.1093/humupd/dmg017
Sex steroids and bone: current perspectives
Juan Balasch
Institut ClõÂnic of Gynecology, Obstetrics and Neonatology, Faculty of Medicine-University of Barcelona, Hospital ClõÂnic-Institut
d'Investigacions BiomeÁdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain
Address for correspondence: Institut ClõÂnic of Gynecology and Obstetrics, Hospital ClõÂnic, C/Casanova 143, 08036-Barcelona, Spain.
E-mail: [email protected]
Although the process of bone remodelling or its control has not yet been fully elucidated there is, at present, suf®-
cient information available to conclude that ovarian steroids (estrogens, androgens, progesterone) play an essential
Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

role in skeletal homeostasis. The mechanism of action of sex steroids on the skeleton is still not entirely clear, but it
has traditionally included indirect effects on systemic hormones that regulate calcium balance and a direct recep-
tor-mediated action. More recently, changes in cytokine production within the bone marrow, as well as pro-apopto-
tic and anti-apoptotic effects in the osteoblastic cells, have been proposed as new perspectives on the cellular and
molecular mechanisms by which sex steroids in¯uence adult bone homeostasis. Mechanical loading, when combined
with estrogens or androgens, results in a greater osteogenic response than either condition separately. Women are
especially at risk for osteoporosis if they have had a premature or surgical menopause and have not received hor-
mone replacement therapy (HRT). Other reproductive factors that can help to identify women with osteopenia and
emphasize the role of sex steroids in preserving bone mass in premenopausal women include: age at menarche, men-
strual history and irregularities (including those associated with excessive exercise), age at menopause, previous hys-
terectomy, hyperprolactinaemia, anorexia nervosa, scoliosis, ovarian dysgenesis, pregnancy and lactation, and
pharmacological ovarian suppression. The prevention of osteoporosis starts with the onset of the menarche. A com-
bination of exercise, appropriate nutrition and a healthy lifestyle all maximize bone mineral accrual and result in
optimal peak bone mass; normal ovarian function is essential to this process. Unfortunately, many women actually
become aware of the need for osteoporosis prevention much later in life, usually after they have already become
menopausal. HRT, however, has important limitations for prevention of fractures in post-menopausal women.
Future perspectives for treatment of osteoporosis include androgen therapy and anabolic agents. Speci®cally, syn-
thetic ligands of the estrogen receptor that can evoke the non-genotrophic but not the genotrophic signal of the
receptor may be bone anabolic agents, as opposed to natural estrogens or selective estrogen receptor modulators
that are anti-resorptive agents. The same ligands may circumvent the side effects associated with conventional
HRT.
Key words: bone homeostasis/bone remodelling/cytokines/hormone replacement therapy/osteoporosis/ovarian steroids
Introduction osteoblast function, and this is still considered today to be a

potential mechanism of estrogen action on bone (Lindsay, 1995).
It was Fuller Albright who, in the 1940s in a series of elegant In addition, Albright reported that loss of androgens in males as a
clinical descriptions, ®rst highlighted the adverse effects of sex result of either chemical or surgical castration or an age-associated
steroid de®ciency on bone (Albright et al., 1940; Albright and decline of androgen levelsÐalbeit not as universal or abrupt as
Reifenstein, 1947). In his original observations, Albright pointed menopauseÐhad the same adverse effect on the skeleton as
out that the prevalence of ovariectomy among osteoporotic women estrogen (Albright, 1947).
was higher than expected, and almost invariably in these women Post-menopausal and ovariectomized women have lower
the surgery had been performed at an age younger than the average circulatory levels of a variety of sex steroids, in addition to
age of the natural menopause. He also showed that the negative estrogens (Lindsay, 1995). Androgens are the most abundant
calcium balance characteristic of osteoporotic post-menopausal circulating sex steroids in both men and women. In women,
women was reversed by estrogen administration. Thus, Albright androgens circulate in the concentration range of nanomolar to
postulated that estrogen in some unknown fashion stimulated micromolar; this contrasts with the estrogens, which have
ã European Society of Human Reproduction and Embryology 207

J.Balasch
circulating concentrations in the picomolar range. Androgens are term HRT is associated with increased risk of breast cancer and the
obligatory precursors in the biosynthesis of estrogens, and in the risk increases with duration of treatment (Cauley et al., 1995;
female are secreted by the adrenal gland and the ovary. Androgens Collaborative Group on Hormonal Factors in Breast Cancer, 1997;
are formed peripherally, particularly from dehydroepiandrosterone Stampfer and Hankinson, 1997; Dixon, 2001; Marcus, 2002). Even
sulphate (Burger, 2002a). The concept of an androgen de®ciency a small increase in relative risk has a large impact on the number of
syndrome in women is a relatively old one, although it has women developing breast cancer because of the high baseline rates
attracted a substantial increase in attention during recent years (Hulka, 1994).
(Burger, 2002b). There are other treatment options to prevent osteoporosis. These
The premenopausal ovary produces signi®cant amounts of options extend far beyond HRT and include lifestyle and dietary
progesterone during the luteal phase of each cycle. In a review of changes such as increasing weight-bearing activity, enhancing
the literature (Prior, 1990) it was indicated that progesterone calcium and vitamin D intake, as well as incorporating pharma-
appears to act directly on bone remodelling and may play a role in cological agents such as the bisphosphonates and selective
the coupling of bone resorption with bone formation. This estrogen receptor modulators (SERMs) (Notelovitz, 1988; 1993;
observation, and the fact that the addition of a progestogen to Bassey, 1995; Marcus, 2002; Pinkerton and Santen, 2002).
estrogen therapy is essential to endometrial protection for post- Regrettably, the different available options are recommended by
menopausal women who have not had a hysterectomy, raises the very few physicians, which suggests that many women are not
question about the role of progestins on bone mass preservation. aware of all treatment options (Randall, 1993).
Although the process of bone remodelling or its control are not Therefore, the aims of the present report were to: (i) analyse the

yet completely understood, there is suf®cient information avail- effects of sex steroids on bone mass; (ii) illustrate those
able to conclude that sex steroids play an important role in skeletal reproductive factors potentially associated with osteopenia; and
homeostasis (Compston, 1990; Lindsay, 1995). The loss of sex (iii) analyse potential therapeutic implications of the relationship
steroids from the ovary, mainly before the age of menopause, between sex steroids and bone mass, stressing limitations of
causes a net loss of bone tissue. When provided to estrogen- universal HRT for prevention of post-menopausal osteoporosis
deprived women, estrogen reverses many of the effects of lost and analysing future therapeutic perspectives. The MEDLINE
ovarian function. Thus, it has been suggested that menopausal database was used to identify relevant English-language articles on
women should use hormone supplements (hormone replacement these topics published between 1980 and August 2002. The
therapy; HRT) for the rest of their lives in order to prevent a medical subject heading terms used for searching included those
variety of ills, including osteoporosis and fractures (Rekers, 1991; for the main topics in this review: sex steroids, estrogens,
Bilezikian and Silverberg, 1992; Bush, 1992; Burkman et al., androgens, progestins, bone, bone mass; risk factors, reproductive
2001). However, the following facts should be considered in this factors, osteopenia; hormone replacement therapy, osteoporosis,
respect: fractures. Additional relevant citations were identi®ed by review-
Ovarian function is an important, but not the only, determinant ing references from retrieved articles and by consulting relevant
of skeletal status among ageing women (Par®tt, 1987; Khosla et al., books on osteoporosis published during the past 10 years.
1995). In fact, early menopause is associated with a self-limiting
loss of bone which does not progress further until ageing exerts its
effect. Thus, a 70-year-old woman has lost 11% of her bone due to Effects of sex steroids on bone mass
menopause and 18% as a function of age; thereafter, the age- The capability of measuring bone mass, using non-invasive
related function is dominant. The main conclusion is that estrogen techniques, and the development of biochemical markers to assess
de®ciency represents one of the multiple causes of involutional bone turnover allowed a more detailed evaluation to be made of
osteoporosis, and its effect is largely limited to bone loss in women the relationship between sex steroids and the skeleton (Delmas,
during the ®rst 5 years after menopause. Other aspects of age- 1995; Johnston and Melton, 1995).
related bone loss (rather than sex steroid de®ciency) dominate the
older years and thus, the signi®cance of early menopause as a risk Estrogens and bone
factor for osteoporosis has been overstated (Nordin et al., 1990; A large number of studies have been conducted to evaluate the
Orwoll and Nelson, 1999; Riggs et al., 2002).
effect of estrogen administration on bone mass in post-
Although all post-menopausal women are estrogen-de®cient,
menopausal women. In general, all studies have drawn similar
osteoporosis develops in only 10 to 20% of them (Khosla et al.,
conclusions. Estrogen intervention reduces the bone remodel-
1995).
Menopausal women are indeed de®cient in endogenous
ling that is seen across menopause, and thus reduces the rate of
estrogens, relative to premenopausal women, if the focus is solely loss of skeletal tissue (Lindsay, 1995).
on the role of endogenous estrogens as preventives of fractures. If All estrogens (both natural and synthetic, and whether
the focus is shifted to the role of endogenous estrogens in the administered by mouth, percutaneously, subcutaneously or
aetiology of breast cancer, ovarian cancer, endometrial cancer and transdermally as appropriate) appear capable of inhibiting bone
uterine ®broids, the premenopausal women are `hyperestrogenic' loss, provided that adequate doses are administered and
and post-menopausal women have a more desirable level adequate serum levels are obtained (Christiansen, 1994;
(Rosenberg, 1993; Hulka and Stark, 1995; Rodriguez et al., Lindsay, 1995; Riggs and Khosla, 1995). Effective doses of
2001; Noller, 2002; Wang and Arnold, 2002). conjugated equine estrogens are about 0.625 mg/day (no
For protection against fractures, estrogen should be initiated additional effect is seen at double this daily dose; Lindsay,
soon after menopause and continued inde®nitely, but this long- 1987; 1995), with comparable doses required for other
208
Sex steroids and bone
estrogens (1±2 mg/day for micronized estradiol). For oral factor, tumour necrosis factor and nitric oxide), prostaglandins
synthetic estrogens the effective doses are as low as 0.02 mg/ (mainly prostaglandin E2) and growth factors (mainly insulin-
day (ethinyl estradiol). For transdermal estrogen, a 0.05 mg like growth factor 1) have been shown to affect bone cells
patch applied every 3.5 days appears to reduce bone loss. in vitro (Compston, 1990; Lindsay, 1995; Riggs et al., 2002).
The effects persist for as long as therapy is provided. ManyÐthough not allÐof these cytokines and growth factors
Moreover, the earlier the treatment is begun, the more are produced by bone cells, and their secretion into the bone
bene®cial is the effect, as bone which has already been lost matrix provides a potential mechanism for the regulation of
cannot be replaced to any signi®cant extent (Lindsay, 1987; bone remodelling (resorption/formation). The results of differ-
1995). Therefore, estrogenÐlike other anti-resorptive agents ent studies have raised the possibility that estrogen may act
(e.g. calcium and bisphosphonates)Ðprimarily prevents fur- directly on cells in the bone microenvironment rather than, or
ther loss of bone mass and thus might be expected to be more in addition to, bone cells themselves (Schot and Schuurs, 1990;
effective when used for prevention (i.e. in early post- Compston, 1994; Lindsay, 1995; Riggs et al., 2002).
menopausal women). However, when estrogen therapy is Alterations in cytokine production within the bone marrow
stopped, bone loss recurs at a rate similar to that after would have signi®cant paracrine or autocrine effects on the
oophorectomy (Notelovitz, 1993; Lindsay, 1995; Marcus, cells responsible for remodelling the skeleton. However, the
2002). For the long-term preservation of bone mineral density, precise role for any of the reported changes within bone is not
continuous estrogen therapy is needed; even 10 years of clear; neither has any cause-and-effect relationship been

treatment after menopause will have little residual effect on demonstrated between the alterations observed in cell or tissue
bone density among women aged 75 years or more, who have culture and estrogen de®ciency bone loss (Lindsay, 1995).
the highest risk of fracture (Felson et al., 1993; Cauley et al., Despite that, the autocrine and paracrine basis for sex steroid
1995; Marcus, 2002). This is of paramount importance action on bone cells has during the past few years come into
considering that exogenous hormones are considered a risk sharper focus, and how to weigh the importance of the various
factor for breast cancer (Hulka, 1994; Dixon, 2001). Of greater cytokines and quantify their complex interactions in mediating
impression is the fact that older women (aged >70 years) the sex steroid effects are subjects for further research (Riggs
continue to lose bone density in spite of estrogen therapy, et al., 2002).
which suggests that the causes of bone loss at these ages are Recently, an interesting new perspective on the cellular and
more complex than during the early post-menopausal period, molecular mechanisms by which sex steroids (estrogens and
and that estrogen alone is incapable of ensuring skeletal androgens) in¯uence adult bone homeostasis and, by exten-
integrity (Orwoll and Nelson, 1999). sion, the pathogenetic mechanisms responsible for the devel-
Despite all the above evidence, the mechanism of action of opment of osteoporosis following sex steroid de®ciency, have
estrogen on the skeleton is still not entirely clear. The been developed by Manolagas' group (Manolagas et al., 2002).
consensus is that the principal action of estrogen is inhibition Until the 1990s, investigations into the pathophysiology of
of bone resorption, although several studies have pointed to a osteoporosis had focused on the functions of differentiated
positive effect of estrogen on bone formation. Speci®c cells: the rates and/or total amounts of bone resorbed or formed
receptors for estrogens (in low concentrations) have been by individual osteoclasts or osteoblasts. However, these cells
identi®ed in cells of osteoblast lineage (Eriksen et al., 1988; have relatively short active lifespans that are within the range
Komm et al., 1988); hence, the effects of estrogen might result of lifespans of other cells originating in the bone marrow. Such
from either direct, receptor-mediated actions or from indirect cells must be continually replaced: the number present depends
actions on systemic hormones (Compston, 1990; Schot and on the birth rate, which re¯ects the frequency of cell division of
Schuurs, 1990). the appropriate precursor cell; and also on the lifespan, which
Estrogen may affect the hormones that regulate calcium most likely re¯ects the timing of death by apoptosis. The
balance, such as calcitonin and parathyroid hormone (Lindsay, importance of the distinction between cell number and
1987; Schot and Schuurs, 1990; DeCherney, 1993). However, individual cell function has often been obscured by using the
the identi®cation of estrogen receptors in normal osteoblast- vague term àctivity' (Manolagas, 1999; Manolagas et al.,
like cells invites speculation that estrogen may have a direct 2002). The essential hypothesis tested by Manolagas' group is
effect on bone rather than affect bone indirectly via secondary that the balance between bone resorption and bone formation
effects on other systemic hormones such as calcitonin and depends more on the number of cells carrying out these
parathyroid hormone. Estrogen receptors, however, are present processes than on their individual capacities.
only in low concentrations (much lower than in reproductive Results obtained from studies conducted by Manolagas'
tissues), and their signi®cance has yet to be demonstrated group during the past 10 years have con®rmed that estrogens
(DeCherney, 1993; Riggs et al., 2002). and androgens decrease the number of bone remodelling cycles
An attractive hypothesis is that estrogen effects are cell- by attenuating the birth rate of osteoclasts and osteoblasts from
mediated, the release of cytokines and growth factors being their respective progenitors (for a review, see Manolagas et al.,
altered in response to estrogenic stimulation (Compston, 1994; 2002). These effects result, in part, from the transcriptional
Lindsay, 1995; Riggs et al., 2002). A number of cytokines regulation of genes responsible for osteoclastogenesis and
(mainly interleukins 1 and 6, macrophage-colony-stimulating mesenchymal cell replication and/or differentiation, and are
209
J.Balasch
exerted through interactions of the ligand-activated receptors (Notelovitz, 2002a). This hypothesis is supported by the results
with other transcription factors. However, increased remodel- of a study conducted in young men which showed an increase
ling alone cannot explain why a loss of sex steroids tilts the in the biochemical markers of bone formation after resistance
balance of resorption and formation in favour of the former. exercise training, with a transient suppression of bone
Estrogens and androgens also exert effects on the lifespan of resorption markers (Fujimura et al., 1997).
mature bone cells: pro-apoptotic effects on osteoclasts, but Approximately 4% of muscle mass is lost during the ®rst 3
anti-apoptotic effects on osteoblasts and osteocytes. These years after menopause, and this is associated with a signi®cant
latter effects stem from a heretofore unexpected function of the decline in muscle strength. In men, muscle strength is
classical `nuclear' sex steroid receptors outside the nucleus, preserved until the age of 60 years, and reaches levels found
and result from the activation of a Src/Shc/extracellular signal- in menopausal women at about 75 years of ageÐa factor that
regulated kinase signal transduction pathway, probably within might explain the greater tendency for falls in older women
pre-assembled scaffolds called caveolae (Kousteni et al., (Notelovitz, 2002a).
2001). Remarkably, either estrogen receptor alpha or beta or
the androgen receptor can transmit anti-apoptotic signals with Androgens and bone
similar ef®ciency, irrespective of whether the ligand is an Androgens have a profound effect on the physiology of bone
estrogen or an androgen. Even more importantly, these non- and muscle in women, as androgens modulate the bone-
genotrophic, sex non-speci®c actions are mediated by the remodelling cycle by direct androgenic activity and transform-

ligand-binding domain of the receptor and can be functionally ation into estrogens (estrone is formed by the aromatization of
dissociated from transcriptional activity with synthetic ligands androstenedione and estradiol by the aromatization of testos-
(Manolagas et al., 2002). terone).
Overall, the above lines of evidence strongly suggest that, in Androgen de®ciency in malesÐlike estrogen de®ciency in
sex steroid de®ciency, loss of transcriptional effects may be femalesÐcan cause osteoporosis, and androgen-related ana-
responsible for the increased osteoclastogenesis and osteo- bolic steroids are sometimes used in the treatment this
blastogenesis, and thereby the increased rate of bone remodel- condition. However, their use is limited by adverse side effects
ling. Loss of non-genotrophic anti-apoptotic effects on mature related to blood lipids and virilization (Rodan and Rodan,
osteoblasts and osteocytes, in combination with an opposite 1995). Clinical evidence of androgen effects on bone mass
effect on the lifespan of mature osteoclasts, may be responsible comes from women with hirsutism, polycystic ovarian disease
for the imbalance between formation and resorption and the and secreting ovarian tumours who often have increased bone
progressive loss of bone mass (Manolagas et al., 2002). mineral density (Gregoriou et al., 2000; Zborowski et al., 2000;
Less well recognized in the bone-remodelling cycle is the Castelo-Branco et al., 2003). Also, the combination of
role of mechanical strain and its effect on the modulating androgens and estrogen replacement therapy in post-meno-
activity of osteocytes. However, the importance of this pausal women increases bone mass to a greater extent than
interaction of sex steroid action and biomechanical forces in estrogen alone; its effects on lipids, however, are clearly
determining the level of bone mass has recently been adverse (Castelo-Branco et al., 2000). In contrast, the use of a
emphasized (Notelovitz, 2002a; Riggs et al., 2002). There gonadotrophin-releasing hormone (GnRH) agonist associated
are clear associations among muscle mass, muscle strength and with low-dose add-back HRT as a prototype contraceptive to
bone density. Mechanical loadingÐcombined exercise and prevent breast cancer was associated with osteopenia because
gravityÐrather than weight-dependent gravity exerts a higher of the inhibition of ovarian androgen production (Spicer et al.,
stimulus on bone formation. Biomechanical strain, and espe- 1993; Pike and Spicer, 2000).
cially that induced by muscle contraction, encourages the Androgen receptors are found in all three bone cells:
activation of bone-forming sites on periostal bone surfaces. osteoblasts, osteoclasts and osteocytes, but are mainly
These changes in bone remodelling will adjust bone mass and expressed in osteoblasts. Speci®c binding receptors for the
distribution to a level that is appropriate for the ambient following androgens have been identi®ed: testosterone,
biomechanical forces. This new bone formation is mediated dihydrotestosterone and dehydroepiandrosterone. Androgens
through the synthesis and activity of both prostaglandin E2 and stimulate osteoblast proliferation, enhance osteoblast differen-
nitric oxide produced by bone surface osteocytes. tiation and the synthesis of extracellular matrix proteins, and
Prostaglandin E2 and nitric oxide are responsible for osteogen- stimulate mineralization (Notelovitz, 2002a;b). Likewise,
esis by modifying the activity of osteoblasts (stimulated by androgens in¯uence bone cell function through their effect
prostaglandin E2) and osteoclasts (inhibited by nitric oxide) on local factors that control the bone cells' microenvironment;
(Notelovitz, 2002a; Riggs et al., 2002). moreover, androgens also have pro-apoptotic effects on
Mechanical loading, when combined with estrogen, results osteoclasts, but anti-apoptotic effects on osteoblasts and
in a greater osteogenic response than does either condition osteocytes, and increase muscle mass and strength
separately. It is the interaction between the two that requires (Manolagas et al., 2002; Notelovitz, 2002a; Riggs et al.,
understanding and which holds the key to success (Notelovitz, 2002). The latter results in an improvement in physical activity,
2002a). This is probably the result of estrogen's anti-resorptive and leads to the activation of bone-forming sites and the
effect and of the stimulation of bone formation with exercise stimulation of bone formation-modulating cells, the osteocytes.
210
These effects of androgens contribute to their action on mineral density when compared with estrogen-only-treated
enhancing bone formation. However, androgen may also act at patients (Speroff et al., 1996). In contrast, micronized proges-
the tissue level by reducing bone resorption (Riggs et al., terone and the less androgenic MPA did not contribute
2002). signi®cantly to the bone-protective effects of estrogen-only
therapy (The Writing Group for the PEPI Trial, 1996).
Progestins and bone In contrast to estrogens, progestins are thought mainly to
The role of progestins in preventing bone loss is less well in¯uence bone formation in the remodelling process.
understood than that of estrogen. There is general agreement, Progesterone receptors have been identi®ed in primary cultures
however, that the synthetic nortestosterone-derived C19 pro- of human osteoblasts and osteoclasts (Slootweg et al., 1992;
gestins with androgenic properties (e.g. norethindrone and Wei et al., 1993). However, it is not clear whether the activities
norethindrone acetate), at doses higher than those needed for of a progestogen on bone can be ascribed to its progestational or
HRT, do have bene®cial effects on bone density. Thus, they can other intrinsic hormonal actions. Besides their progestational
increase bone density in women with post-menopausal osteo- activities, progestogens exert androgenic, corticosteroidal and
porosis (Horowitz et al., 1993) and ameliorate the effects of even estrogenic activities. Furthermore, from a practical point
estrogen de®ciency in younger women receiving GnRH of view, of more interest and importance is the question of
agonists (Abdalla et al., 1985; Riis et al., 1990; Eldred et al., whether progestogens can interact with the estrogen effects on
1992). However, data regarding the effects on bone of the the skeleton. In this regard it is interesting to note the results of a

progesterone-related C21 progestins such as medroxyprogester- study (Slootweg et al., 1992) where the authors investigated the
one acetate (MPA) are contradictory. Thus, it has been reported action of a `pure progestogen' (i.e. binding only to the
that in premenopausal women with luteal phase de®ciency, progesterone receptor), Org 2058, in SaOS-2 human osteogenic
MPA (10 mg/day; 10 days/cycle) could signi®cantly improve osteosarcoma cells. The pure progestogen had no effect on cell
spinal bone density (Prior et al., 1994). In contrast, 20 mg/day proliferation when added alone. However, in combination with
of that progestin were unable to arrest spinal bone loss in post- 17-b-estradiol, it had a highly synergistic action on SaOS-2 cell
menopausal women (Gallagher et al., 1991). Also, premeno- proliferation. The same effect was observed in primary rat
pausal women using parenteral depot MPA (DMPA) for osteoblasts, showing that non-transformed cells react similarly
contraception or prescribed high-dose (50 mg/day) MPA for and thus, that this synergism is a general phenomenon in
gynaecological disorders had signi®cantly reduced bone dens- osteoblastic cells. The effects of estrogen and progestogen on
ity (Cundy et al., 1991; 1996). The variable ®ndings in these bone cells suggested the presence of receptors for these
studies clearly suggest that the effects of MPA on bone density compounds in SaOS-2 cells. The authors therefore performed
may vary according to the dose used and the estrogen status of classical biochemical receptor studies to demonstrate the
the user. When given in doses suf®cient to induce hypogonad- presence of sex steroid receptors in these cells. Scatchard
ism, MPA use is associated with a signi®cant and rapid loss of analysis revealed a signi®cant number of high-af®nity estrogen
bone density from the lumbar spine. Traditionally, it has been and progestagen receptors in the SaOS-2 cells, thus indicating
accepted that this bone loss is the consequence of estrogen that the effects of these steroids are mediated via the normal
de®ciency and occurs despite a gain in body weight, although it route of steroid receptors (Slootweg et al., 1992). In this regard,
appears to be at least partially reversible (Cundy et al., 1996). it has also been shown that some synthetic progestagens exert
Recently, however, it has been argued that bone loss associated estrogenic effects through the activation of estrogen (but not
with MPA administration is caused by decreased osteoblast progesterone) receptors, and so can have a dual effect on
differentiation as a result of medroxyprogesterone occupying estrogen target tissues, either to stimulate or differentiate cells
the glucocorticoid receptor, since increasing glucocorticoid (Jeng et al., 1992; Jordan et al., 1993). These investigations
receptor occupancy beyond that reached at optimal glucocorti- lead to the conclusion that some estrogen-like effects could be
coid concentrations attenuates osteoblast differentiation (Ishida produced by synthetic progestins. Interestingly, 19-nortestos-
and Heersche, 2002). If this hypothesis is con®rmed, then terone derivativesÐbut not MPAÐwere found to have marked
progestins with no glucocorticoid activity would be a better estrogenic properties. On the other hand, experimental studies
choice for progestagen therapy in order to achieve more in rats have shown that estrogens may play an important role in
bene®cial effects on bone metabolism. maintaining bone mass by up-regulating progesterone receptor
While the majority of the data accumulated on estrogens and levels in a class of osteoprogenitor cells responsive to
skeletal status have dealt with estrogen alone, the requirement progesterone (Ishida and Heersche, 1999). Overall, these data
to consider combined HRT for all women with a uterus provide a cellular basis for the clinically recognized positive
demands that the effects of progestins be examined in this effect of estrogen and progestogen combinations on bone mass.
context. In addition, because of their other metabolic effects
there is agreement that progestins should not be used them-
selves for osteoporosis prevention or treatment (Lobo and Risk factors of osteopenia related to sex steroids
Speroff, 1994). Clinical trials have shown that post-meno- The prevention of osteoporosis is of utmost importance in reducing
pausal women receiving norethindrone acetate in association the costs related to this disease. In women, bone loss prior to
with estrogen displayed a signi®cant increase in their bone menopause may contribute to later risk of fracture due to
211
J.Balasch
osteoporosis. Therefore, an important prevention strategy is to associated with lower bone mineral content (Sowers et al.,
optimize and maintain premenopausal bone mineral status (Tudor- 1998). Defective luteal phases have been linked with spinal
Locke and McColl, 2000). bone density losses in some (Prior, 1990), but not all (Waller
During the early years of life, adolescence and early adulthood, et al., 1996; DeSouza et al., 1997), studies. Bone loss in women
bone mass increases until, at a point in time which remains to be with anovulatory menses is estimated at 4.2% per year
clearly de®ned, the peak bone mass is attained. Most of the bone (Notelovitz, 1993). Bone loss has also been associated with
mass will be accumulated by late adolescence after cessation of low levels of androgens in premenopausal women (Slemenda
longitudinal growth (Tudor-Locke and McColl, 2000), but peak et al., 1996).
bone mass attainment probably occurs during the third decade of A history of menstrual irregularities has been consistently
life. This may be followed by a period of relative stability until the
associated with lower bone mineral density in premenopausal
point at which age-related bone loss begins, prior to menopause, in
women. A reduced bone mineral content has been shown to be
the fourth decade or early in the ®fth decade (Compston, 1990).
related to duration of amenorrhoea and severity of estrogen
Peak adult bone mass (the actual amount of bone in the skeleton) is
a major factor determining the risk of fracture. Thus, if a woman
de®ciency, regardless of the underlying cause of the irregular-
with a good peak adult bone mass at menopause loses the average ity (Tudor-Locke and McColl, 2000). It has been estimated that
20% of bone mass, the loss is not suf®cient to increase the risk of women who missed <50% of their expected menses had a
fracture to dangerous levels. By contrast, if a woman with a low vertebral bone mass that was 88% of their eumenorrhoeic
peak adult bone mass at menopause loses only half as much peers, and those who missed >50% of their menses had values

boneÐabout 10%Ðthe risk of fracture increases rapidly that were 69% of those of normally menstruating women.
(Stevenson, 1990). Therefore, there exists the potential to provide Serum estradiol concentrations were lower and subnormal in
premenopausal women with the information necessary to enter both groups of oligomenorrhoeic women (Notelovitz, 1993).
menopause with an individually optimized bone mineral status. Women who experience periods of regularity, interspersed
In the past, health education strategies have focused on the with times of oligomenorrhoea and/or amenorrhoea, are
needs of post-menopausal women, and the potential of prevention intermediate to these other two groups with respect to bone
strategies targeted at premenopausal women has largely been mineral density, suggesting some protection with infrequent
ignored. However, suf®cient evidence-based knowledge is now regularity (Tudor-Locke and McColl, 2000).
available regarding risk factors for reduced mineral status in this The deleterious effects of amenorrhoea on bone are
population, to provide a guide for behavioural changes that are immediate, and therefore early recognition and intervention
conducive to good bone health (Tudor-Locke and McColl, 2000). are necessary. Increases in bone mineral density are
The risk factors for osteopenia in relation to ovarian sex steroids associated with resumption of menses, which is the
are discussed in the following sections. treatment goalÐpreferably by treating the underlying
Age at menarche cause. Regardless, it is important to re-establish a regular
estrogen±progesterone cycle with appropriate replacement
Early and continued exposure to endogenous estrogen during a therapy if necessary (Crosignani and Vegetti, 1996; Tudor-
time of accelerated bone growth may potentiate peak bone Locke and McColl, 2000).
mass development in premenopausal women. The onset of the
menarche is thus considered as an important biological Age at menopause
indicator of future bone mass. Several studies of age at
menarche and premenopausal bone have suggested the pres- Delayed menopause, or alternatively an extended reproductive
ence of an inverse relationship: the earlier the onset of period characterized by a longer protective hormonal environ-
menstruation, the greater the individual's subsequent bone ment, is associated with higher bone mineral density and
mass (Notelovitz, 1993; Tudor-Locke and McColl, 2000). A reduced risk of hip fracture in older women (Tudor-Locke and
later age of menarche may indicate underlying hormonal McColl, 2000). In contrast, women are specially at risk for
irregularities, and could therefore be a proxy measure for low osteoporosis and fracture if they have had a premature or
bone density in premenopausal women. Thus, it has been surgical menopause and have not received HRT (Notelovitz,
reported that a later age of menarche was no longer a 1993; Lindsay, 1995).
signi®cant factor for low bone mineral content when controlled
Hysterectomy with ovarian conservation
for nulliparityÐitself a potential proxy indicator for underlying
hormonal abnormalities (Sowers et al., 1992). Premenopausal women who underwent a hysterectomy with
ovarian conservation were found to have bone densities
Menstrual history and irregularities signi®cantly lower than those of normally menstruating
Cyclic disturbance of the menstrual cycle is an important controls (Hreshchyshyn et al., 1988). It is not known
marker of potential osteopenia. However, asymptomatic men- whether this effect is due to removal of the uterus per se,
strual disturbances and no associated amenorrhoea have also to aberrant functioning of the retained ovaries (Notelovitz,
been reported as a risk factor for low bone density in 1993), or to premature loss of ovarian function (Souza
premenopausal women (the so-called `marginal hormonal et al., 1986; Siddle et al., 1987; Cooper and Thorp, 1999).
status') (Sowers and Galuska, 1993). Thus, lower levels of Interestingly, both hysterectomy and tubal ligation have
estradiol in eumenorrhoeic premenopausal women have been been reported to protect against ovarian cancer, even in
212
carriers of BRCA1 and BRCA2 mutations (Hankinson et al., Scoliosis

1993; Rosenblatt et al., 1996; Narod et al., 2001). Although the cause of scoliosis is not known and is associated
with a strong genetic tendency, it has been related to
Exercise hypoestrogenism. Adult scoliotic women have a higher inci-
Exercise is generally considered to be bene®cial to the skeleton dence of osteoporosis, and younger women with scoliosis
and, in fact, high bone density is consistently observed in should be regarded as being at greater potential risk for
eumenorrhoeic athletes while intervention studies in non- osteoporosis (Notelovitz, 1993).
athletic premenopausal women indicate that high-load activ-
ities (weight training, running, jumping and stepping) improve Ovarian dysgenesis (Turner's syndrome)
bone mineral status (Tudor-Locke and McColl, 2000). The severe estrogen de®ciency characteristic of Turner's
Excessive exercise, however (mainly running) may be less syndrome is associated with a signi®cant bone mass decrease
bene®cial for at least a subpopulation of women (Marcus et al., ascribable to increased bone turnover, as indicated by
1985). Ballet dancers are also subject to amenorrhoea and histological studies and assays of bone turnover markers.
lower bone mass than average (Warren et al., 1991). Estrogen therapy is followed by a signi®cant bone mass gain
Amenorrhoeic athletes have a lower body weight than is and a return to normal of bone turnover markers, suggesting
found in those who remain eumenorrhoeic, and they also have that it is the estrogen de®ciency rather than the chromosomal
lower circulating estradiol and progesterone levels (Lindsay, abnormality that causes the bone mass de®ciency, although

1995). It seems likely that it is the reduction in sex steroids abnormalities in the renal metabolism of vitamin D have been
which is associated with reduced bone mass and thus, the return reported (Breuil and Euller-Ziegler, 2001). If not treated with
of menses (usually following a reduction in exercise) results in estrogen, patients with ovarian dysgenesis probably begin to
a degree of recovery (Drinkwater et al., 1986). The lower bone lose bone at early ageÐa feature accompanied by postmeno-
mass of athletes who have lost steroids appears to be associated pausal serum levels of bone markers such as alkaline
with a greater incidence of stress fractures, often affecting the phosphatase and urinary hydroxyproline (Lindsay, 1995).
small bones of the feet (Warren et al., 1991). These athletes do
Pregnancy and lactation
not appear to be at increased risk of classical osteoporotic
fractures (Lindsay, 1995). There is no net effect of pregnancy on bone mineral density
(Rosen and Rosenblatt, 2002). There is, at present, insuf®cient
Hyperprolactinaemia evidence to state that parity or age at ®rst pregnancy are risks
factors for reduced bone mineral status in premenopausal
Failure of the gonadal axis due to a prolactin-secreting
women (Tudor-Locke and McColl, 2000).
pituitary tumour results in a signi®cant loss of bone tissue
Secretion of calcium into the milk of lactating women
(Schlechte et al., 1983). That it is, the loss of sex steroids
approximately doubles the daily loss of calcium. Thus,
that is responsible for the low bone mass is suggested by
lactation is associated with loss of bone of 1.5 to 4% in
the normal bone mass of hyperprolactinaemic women with
women who breast-feed for 6 months or more, even in the
intact ovarian function, but similar prolactin serum levels.
presence of a high calcium intake (Rosen and Rosenblatt,
Treatment of the underlying problem with return of ovarian
2002). The bone loss is due to increased bone resorption,
function is also associated with signi®cant gain in bone
probably secondary to the relatively low estrogen levels
mass (Lindsay, 1995).
associated with lactation. However, bone density returns to
baseline levels in the 6 months after weaning. Recovery of
Anorexia nervosa
bone density appears to be linked with resumption of menses
Patients with anorexia nervosa have signi®cantly reduced (Speroff et al., 1999; Tudor-Locke and McColl, 2000). It is
bone mass compared with normal control women, mainly possible that recovery is impaired in women with inadequate
when anorexia begins before puberty. The reduction in calcium intake. Total calcium intake during lactation should be
bone mass may be more than 25% of that in age-matched at least 1500 mg daily. However, calcium supplementation has
individuals and is related to the duration of amenorrhoea. no effect on the calcium content of the breast milk or on bone
In addition to the calorie deprivation and low body weight, loss in lactating women who have normal diets (Speroff et al.,
mechanisms for bone loss in these patients include also 1999). Lactation followed immediately by a second pregnancy
high endogenous cortisol production and suppression of the does not adversely affect recovery of bone density, and there is
hypothalamic pituitary axis (Lindsay, 1995; Tudor-Locke no increased risk of osteoporosis and fractures in post-
and McColl, 2000). There are concerns that incurred bone menopausal women who lactated earlier in life (Tudor-Locke
de®cits may incompletely recover from anorexia nervosa. and McColl, 2000; Rosen and Rosenblatt, 2002).
Bone loss continues with ongoing amenorrhoea. HRT or
oral contraceptives increase bone density, but the best Pharmacological ovarian suppression
remission is associated with natural resumption of menses. Suppression of ovarian function is used mainly for contracep-
Not infrequently, however, recovery is slow and incomplete tion and treatment of endometriosis, and produces varying
(Tudor-Locke and McColl, 2000). degrees of estrogen de®cit. Indeed, prolonged suppression in
213
J.Balasch
premenopausal women may have a signi®cant impact in ceptive use on bone in healthy premenopausal women with
accelerating bone loss. Data in the literature do not support any regular menses.
differences in bone mass in the spine, hip or forearm of normal
MPA and progestin implants
women from those with endometriosis; thus, women with
endometriosis are more likely to be at risk from their medical DMPA is a widely used injectable form of hormonal contra-
therapy rather than any inherent initial osteopenia (Dawood, ception (Kaunitz, 2002). In contraceptive doses, DMPA
1994; Tudor-Locke and McColl, 2000). suppresses ovarian production of estradiol, which suggests a
potential for an increased risk or osteopenia and subsequent
Oral contraceptives
post-menopausal osteoporosis. As discussed above, the impact
Oral contraceptives are commonly recommended for the of premenopausal DMPA use on post-menopausal osteoporosis
treatment of hypoestrogenic amenorrhoea, and it seems clear remains controversial. Thus, although it is considered a safe
that a bone-sparing bene®t of oral contraceptive use is observed and reliable form of contraception and its impact on bone
in this group, though not to the extent seen with a natural mineral density similar to that of lactation (i.e. both lower
resumption of menses (Tudor-Locke and McColl, 2000). ovarian production of estradiol, leading to reversible decreases
However, the effects of oral contraceptives on bone in healthy in bone mineral density), the United States Food and Drug
premenopausal women with regular menses have been con- Administration recommended post-marketing studies to deter-
troversial (Speroff et al., 1999; Tudor-Locke and McColl, mine, prospectively, changes in bone density related to DMPA

2000; Dayal and Barnhart, 2001). Traditionally, it has been use; the results of a long-term study initiated in 1994 are
suggested that oral contraceptives may preserve bone mineral expected in 2003 (Kaunitz, 1998; Tudor-Locke and McColl,
density. In general, however, epidemiological study of the 2000).
effects of oral contraceptives on bone mineral density has been Fluctuating estradiol levels in progestin implant or proges-
limited by evaluation of post-menopausal, former high-dose tin-releasing intra-uterine system users imply that loss of bone
oral contraceptive users. In fact, some retrospective studies density would be unlikely. The potential different effects on
have found a bene®cial effect of oral contraceptives on bone mass of progestin implants or progestin-releasing intra-
premenopausal bone, whilst others have not. Discrepancies uterine systems and DMPA is probably attributable to less
are due to dose, duration of use, characteristics of subject suppression of the pituitary gland by implants, permitting more
populations, including wide age ranges and variable menstrual follicular activity in the ovary and more estrogen production
status, and potential confounders, including calcium intake and (Cromer, 1999; Luukkainen et al., 2001; Meckstroth and
smoking habits. Moreover, not all oral contraceptives are Darney, 2001).
similar, making it dif®cult to generalize their effects (Tudor-
Danazol
Locke and McColl, 2000). Thus, studies re¯ecting modern use
of low-dose products indicate little impact of oral contraceptive Because of its androgenicity, danazol use is associated with
use on bone mass (Mais et al., 1993; Hartard et al., 1997; either signi®cant bone gain or no change in bone mineral
Speroff et al., 1999). density (Hornstein and Barbieri, 1988; Dawood, 1994).
However, measurements of bone density are not as import-
GnRH agonists
ant as the clinical outcome, namely fractures (Speroff et al.,
1999). In a population-based, case±control study of post- The long-term consequences of the hypoestrogenic state
menopausal women who had formerly taken oral contracep- induced by GnRH agonists on calcium metabolism and bone
tives during their premenopausal years, those aged over 40 are of concern (Dawood, 1994; Speroff et al., 1999). For this
years showed signi®cant protection against hip fractures during reason, therapy is usually limited to 6 months to avoid bone
their menopause, while those aged less than 40 years had loss, though even during this time period there can be a 6±8%
smaller and non-signi®cant reductions in hip fracture decrease in bone mineral density. This loss is reversed after
(MichaeÈlsson et al., 1999). However, the available evidence cessation of therapy, but many patients (mainly those of
suggests that any favourable effects on bone density in advanced reproductive age or with lower calcium intakes) have
premenopausal women are not clinically important (Speroff not regained the bone that was lost, as much as one year later
et al., 1999); this suggestion is supported by two large cohort (Orwoll et al., 1994; Speroff et al., 1999). In this respect, it
studies (Cooper et al., 1993; Vessey et al., 1998). In the Royal should be noted there is not only individual variability of bone
College of General Practitioners study, the overall risk of density but also variability in the amount of bone lost by
fractures in ever-users of oral contraception was actually different individuals receiving the same dose of GnRH agonist
slightly increased (Cooper et al., 1993). Similar results were over the same duration, as is the case with bone loss occurring
observed in the Oxford Family Planning Association Study after the menopause (Dawood, 1994).
(Vessey et al., 1998). It is likely that the increased risk re¯ects In recent years, so-called àdd-back' therapy has been used
lifestyle effects among oral contraception users, but thus far to alleviate a lasting impact on the risk of osteopenia associated
there is no evidence of any protective effect against fractures with long-term therapy with GnRH agonists (Deary et al.,
(Speroff et al., 1999). Therefore, prospective studies of longer 2002). The addition of a 19-nortestosterone progestin, a post-
duration are needed to determine the effects of oral contra- menopausal estrogen-progestin programme, tibolone or bi-
214
Figure 1. Schematic representation of the four major interactive factors

determining bone mass in the adult female.
sphosphonate treatment may be effective in reducing bone loss Figure 2. Diagrammatic representation of bone mass accrual in women and
associated with GnRH agonist therapy while maintaining factors in¯uencing the bone mass at different ages. The different factors are
treatment ef®cacy (Edmonds, 1996; Moghissi, 1996; Speroff shown in the boxes on the horizontal axis. Modi®ed from Heaney (1987).
et al., 1999). Interestingly, however, it has been reported that a

daily dose 0.625 mg conjugated estrogen associated with MPA
is not suf®cient to prevent the hypoestrogenic effects of GnRH lifestyle or social habits such as smoking and excessive caffeine,
agonists on skeletal bone, and 1.25 mg of conjugated estrogen alcohol, sodium and animal protein intake (Heaney, 1987;
or the equivalent dosage of other estrogens should be Dawood, 1994; Tudor-Locke and McColl, 2000).
As discussed above, peak bone mass is usually attained by the
prescribed when planning long-term GnRH agonist ovarian
third decade of life, and this maximum accrual of bone mass serves
suppression (Sugimoto et al., 1993).
to put a woman into a more advantageous skeletal balance in
In addition to endometriosis, this add-back therapy is
preparation for the impending accelerated bone loss accompany-
considered to be a breakthrough for the management of a
ing the early post-menopausal years. This emphasizes the para-
number of gynaecological conditions including uterine leio- mount importance of maintaining or increasing bone mass before
myoma, dysfunctional uterine bleeding and premenstrual the decline observed after menopause. Despite genetic control of
syndrome (Studd and Leather, 1996). In contrast with peak bone density, nutrition, exercise and hormonal milieu can
treatment of uterine ®broids, add-back therapy for endome- impact on bone during growth, thereby either reducing or
triosis does not need to be deferred for optimal results (Kiesel increasing the capacity to reach the maximum genetic potential.
et al., 1996). A matter of concern is that these gynaecological To in¯uence bone density, therefore, intervention is required at a
conditions are frequently recurrent diseases and thus further very early stage. Given the dynamic activity of the bone-
treatments may become necessary. Any medical treatment remodelling cycle during puberty, this could be the optimal time
associated with a signi®cant loss of bone mass, and without a to in¯uence bone mass accrual (Stevenson, 1990; Notelovitz,
complete return to baseline several months after treatment is 1993).
stopped, may produce an additional cumulative de®cit in Therefore, the prevention of osteoporosis starts with the onset of
trabecular bone mass with each succeeding treatment. With a the menarche. A combination of exercise (to stimulate new bone
suf®cient treatment-free interval of perhaps more than a year, formation), appropriate nutrition (to mineralize the newly formed
and with adequate calcium intake and elimination of risk osteoids) and a healthy lifestyle (to avoid detrimental effects on
factors for osteoporosis, it is likely that bone loss will be bone formation) all maximize bone mineral accrual and result in
adequately recovered (Dawood, 1994), but in some women optimal peak bone mass. Normal ovarian function (estrogen-
bone density is restored only by 2 years after a 6-month period replete state to modulate the rate of bone loss) is essential to this
of GnRH agonist therapy (Paoletti et al., 1996). Therefore, it process (Figure 2) (Heaney, 1987; Notelovitz, 1993; Dawood,
should be considered that re-treatment with GnRH agonist or 1994). Unfortunately, many doctors rely on HRT only, and most
any other medication producing bone loss after short or women actually become aware of the need for osteoporosis
prevention much later in lifeÐusually after they have already
inadequate post-therapy intervals is likely to cause recurring
become menopausal (Notelovitz, 1993). HRT, however, has
additional residual bone losses, giving rise to an increasing
important limitations for prevention of fractures in post-meno-
de®cit in total bone mass (Dawood, 1994).
pausal women, as discussed next.
Sex steroids and bone: therapeutic implications Limitations of HRT for prevention of fractures
Bone mass in the adult female is determined by four major The ®rst report from the Women's Health Initiative (WHI)
interactive factors: (i) heredity (genes); (ii) exercise (mechanical study has recently been published (Writing Group for the
loading); (iii) nutritional status (intake of calcium and other Women's Health Initiative Investigators, 2002). The WHI
nutrients); and (iv) hormonal status of the person (sex steroids) is the ®rst randomized clinical trial to demonstrate a
(Heaney, 1987; Notelovitz, 1993; Dawood, 1994) (Figure 1). reduction in hip and vertebral fracture risk with HRT use,
Superimposed on the complex interaction of these factors are and the ®ndings are discussed below. Historically, and
215
J.Balasch
before such a study, support for estrogen use to prevent osteoporosis as an indication for estrogen therapy because of
bone loss and fractures in post-menopausal women had the lack of evidence from randomized trials of the effect of
come from three lines of evidence. First, randomized trials estrogen on the risk of fracture''.
have consistently shown that estrogen prevents post-meno- Ideally, clinical practice should be based on evidence from
pausal bone loss. Many short-term studies and some large randomized trials using fracture as the outcome measure.
longer-term studies of HRT and bone mineral density as Such trials have found that bisphosphonates reduce non-
the primary outcome have shown signi®cant ef®cacy vertebral fracture risk as well as vertebral fracture risk whilst
(Speroff et al., 1996; The Writing Group for the PEPI raloxifene and calcitonin reduce vertebral fracture risk, and
Trial, 1996; NIH Consensus Development Panel on these drugs have been approved by the Food and Drug
Osteoporosis Prevention, Diagnosis, and Therapy, 2001). Administration for prevention and treatment of osteoporosis
Second, observational studies consistently suggested that (Grady and Cummings, 2001). Surprisingly, no such large trial
post-menopausal HRT reduces the risk of hip and other has been performed to determine the effect of estrogen on
types of fracture (Grady et al., 1992; Lindsay, 1995). fracture risk in women with osteoporosis (Grady and
Third, evidence was obtained from risk-versus-bene®t Cummings, 2001). The Heart and Estrogen/progestin
studies of HRT for post-menopausal women who are Replacement Study (HERS and HERS II) (Cauley et al.,
considering long-term hormone therapy to prevent disease 2001; Hulley et al., 2002) and the WHI randomized controlled
or to prolong life. According to these studies, HRT would trial (Writing Group for the Women's Health Initiative

increase life expectancy for most menopausal women, since Investigators, 2002), however, have recently examined the
protection against hip fractures and coronary heart disease effect of long-term post-menopausal hormone therapy on
would outweigh the increase in the risk of breast cancer fractures. The HERS study was a randomized, blinded,
(Grady et al., 1992; Burkman et al., 2001). placebo-controlled trial of estrogen plus progestin therapy
Reduced bone density has been considered to be a risk factor after menopause of 4.1 years duration (HERS) and subsequent
for fracture. However, change in bone density is an imperfect open-label observational follow-up for 2.7 years (HERS II).
predictor of reduction in the risk of non-spine fractures (Grady The study included 2763 post-menopausal women with
and Cummings, 2001). Thus, for example, calcium with coronary heart disease; 2321 of these women (93% of those
vitamin D decreases the risk of fracture, despite having little surviving) consented to follow-up in HERS II. The WHI was
or no effect on bone density. In contrast, alendronate (which the ®rst randomized trial to directly address whether estrogen
reduces fracture risk in women with osteoporosis) substantially plus progestin has a favourable or unfavourable effect on
increases bone density but does not reduce the risk of non-spine coronary heart disease incidence and on overall risks and
fractures in women without osteoporosis (Grady and bene®ts in predominantly healthy women. A total of 16 608
Cummings, 2001). women entered the randomized double-blind trial (planned
Although the skeletal bene®ts of HRT for preventing bone duration, 8.5 years), and after a mean of 5.2 years of follow-up
loss can hardly be challenged, one might be wary of published the data and safety monitoring board recommended stopping
evidence that prolonged HRT use unequivocally reduces the the trial because the test statistic for invasive breast cancer
risk of fracture (Reginster et al., 2000). Remarkably, unlike exceeded the stopping boundary for this adverse effect, while
trials of the effects of estrogen replacement on bone mass in the the global index statistic supported risks exceeding bene®ts.
early menopause (often placebo-controlled), the ability of The HERS main trial revealed little difference between the
estrogen to reduce fracture risk has been reported (until very hormone and placebo groups in risk of any type of fracture
recently, as discussed below) only in uncontrolled, frequently (Cauley et al., 2001). Surprisingly, the additional follow-up
retrospective, observational or case-controlled studies (Orwoll experience from HERS II suggests a risk of hip fracture among
and Nelson, 1999; Reeve, 2000; Reginster et al., 2000; Grady women in the hormone therapy group that was higher than that
and Cummings, 2001). Observational studies, however, are in the placebo group (relative hazard, 1.61; 95% CI, 0.98±2.66;
susceptible to selection bias and confounding. This may be P = 0.06). The WHI (Writing Group for the Women's Health
especially true of studies of HRT because users tend to be Initiative Investigators, 2002) was the ®rst trial with data
healthier, wealthier and more active than non-users (Orwoll supporting the ability of post-menopausal hormones to prevent
and Nelson, 1999; Reeve, 2000; Grady and Cummings, 2001; fractures at the hip, vertebrae and other sites (hazard ratio for
Laine, 2002; Vastag, 2002). Therefore, an important part of the all types of fractures, 0.76; 95% CI, 0.69±0.85). Discrepancies
protective effect associated with taking estrogen is mediated between studies may be explained by the fact that women
via other factors (exercise, diet and others). For this reason, included were not selected for osteoporosis and may thus not
concern has been raised as to the preferential treatment given to be well suited to revealing the effects of fracture-prevention
HRT products compared with other currently developed treatments. Clinical trials of bisphosphonates have found an
medications, where evidence of an anti-fracture ef®cacy at effect on the risk of fracture in women with osteoporosis, but
the level of the spine or hip is requested (Reginster et al., not in women with normal bone density (Cummings et al.,
2000). As recently stressed (McNagny and Wenger, 1998; McClung et al., 2001). Overall, however, those studies
2002),¼``clinicians should alert women that, in 1999, the cast doubt on the usefulness of HRT as a universal approach to
Food and Drug Administration removed the treatment of prevent fractures in post-menopausal women. An evaluation of
216
incident fracture rates in the Study of Osteoporotic Fractures of estrogen therapy may have underestimated the risk of breast
population further illustrates the limitations of HRT. In women cancer and overestimated the prevention of fractures''
who reported taking estrogen continuously since the meno- (Hemminki and Sihvo, 1993). The results of that study
pause (an average of 25 years), the rate of bone loss was noted supported the tenet that, until the 1990s, physicians in the
to accelerate with increasing age, and those on estrogen therapy USA tended to select healthier women for long-term hormone
continued to lose bone density, as did those without estrogen therapy, and women predisposed to osteoporosis or with an
(Ensrud et al., 1995). Furthermore, as reported very recently, increased risk of breast cancer were excluded from therapy.
almost 20% experienced a non-traumatic, non-vertebral frac- This further limits the potential value attributed to HRT in
ture during a 10-year follow-up period (Nelson et al., 2002a). early studies.
This was fully two-thirds the number observed in women who The studies that formed the foundation of the 1992
had never taken estrogen. The extent to which estrogen was guidelines (American College of Physicians, 1992; Grady
associated with lower rates of vertebral and hip fracture was et al., 1992) were biased because most of them did not
similar. Clearly, many women who take estrogen will ultim- adequately account for socioeconomic status, and we now
ately suffer osteoporotic fractures (Orwoll and Nelson, 1999). know that the women who take estrogen are different from
Therefore, low bone mass and fractures remain serious threats those women who do not: they are healthier; they are wealthier;
in older post-menopausal women, even in the presence of and they have a better health pro®le. Thus, rather than HRT
hormone replacement. Thus, as recently stressed (Grady and keeping women healthy, healthy women were taking HRT

Cummings, 2001), perhaps the important clinical question now (Laine, 2002; Vastag, 2002).
is not whether estrogen reduces risk of fracture, but whether it Recent controlled trials (such as HERS, WHI, Estrogen
reduces risk as much as other proven useful treatments such as Replacement and Atherosclerosis Trial, Women's Estrogen for
bisphosphonates. Stroke Trial, Papworth Hormone Replacement Therapy
Finally, since women in their 50s who do not have Atherosclerosis Study) have brought risks and bene®ts of
osteoporosis have a relatively low risk of fracture, the bene®t HRT under scrutiny, and investigators renounced their earlier
of long-term treatment with estrogen to prevent bone loss and recommendations to prescribe such therapy for prevention in
fractures may not exceed the risks. Previous guidelines for post-menopausal women (Friedrich, 2001; Vastag, 2002).
counselling post-menopausal women about preventive hor- Falling in line with the evidence-based medicine trend
mone therapy (American College of Physicians, 1992) were (Herrington, 1999; Cauley et al., 2001; Hulley et al., 2002;
based on estimates of the potential risks and bene®ts coming Humphrey et al., 2002; Nelson et al., 2002b; Writing Group for
from results of observational studies (Grady et al., 1992). the Women's Health Initiative Investigators, 2002), an inter-
According to those estimates a 50-year-old healthy white national team of women's health experts is discouraging the
woman has a 15% lifetime probability of suffering a hip use of HRT for many post-menopausal conditions. Coronary
fracture (median age at ®rst hip fracture, 79 years) and a 1.5% heart disease, fractures, depression and urinary incontinenceÐ
probability of dying of a hip fracture. If treated with long-term all of which have in the past been cited as prime reasons to
HRT, the lifetime probability of hip fractures is reduced to initiate HRTÐare losing favour as valid indications for its use,
12%. However, the estimated lifetime probability of breast as evidence from high-quality clinical trials accumulates
cancer is increased substantially and, even considering a (Vastag, 2002). Although short-term use (<5 years) of HRT
bene®cial effect on the cardiovascular system, the overall with combined estrogen-progestin may help women cope with
bene®t is negligible: the estimated increase in life expectancy symptoms of the menopause, when used for longer periods this
is only 0.1 years. For a woman who is at increased risk for hip therapy may cause net harm. Whilst event rates are low,
fracture due to such factors as low bone mineral density, the possible reductions in the risk of fracture and colorectal cancer
overall bene®t is again small (estimated increase in life are outweighed by increases in the risk of coronary heart
expectancy of 0.2 years). In women at increased risk for breast disease, stroke, pulmonary embolism, breast cancer and biliary
cancer, treatment with combination hormones could result in a tract surgery (Simon et al., 2001; Nelson et al., 2002b; Writing
substantial increase in lifetime probability of breast cancer and Group for the Women's Health Initiative Investigators, 2002).
a slight reduction in life expectancy (American College of Increased risk of ovarian cancer is also of major concern
Physicians, 1992; Grady et al., 1992). (Lacey et al., 2002; Riman et al., 2002a;b; Rodriguez et al.,
In patients at increased risk for hip fracture, the lifetime 2002). This implies a dramatic shift on best clinical practices
probability of hip fracture does not decrease as much as for treating patients during menopause. This new perspective is
expected because the additional year of life gained occurs, on summarized in the evidence-based recommendations from the
the average, at an advanced age when the incidence of hip International Position Paper on Women's Health and
fracture is very high. On the other hand, breast cancer is a more Menopause: A Comprehensive Approach, published by the
important cause of illness and death at younger ages than either National Heart, Lung, and Blood Institute (NHLBI), National
cardiovascular disease or osteoporotic fractures (Grady et al., Institutes of Health (NIH). The International Position Paper is
1992; Rosenberg, 1993). In this context, it should be noted that based on extensive international review and evaluation of the
a review on potential selection bias in post-menopausal scienti®c evidence for current clinical practices as presented in
hormone therapy concluded that ``surveys on the health impact the published literature. Further information is available at the
217
J.Balasch
Of®ce of Women's Health Research Web site: (http:// and progressively from early adulthood until old age. The impact
www.nhlbi.nih.gov/health/prof/heart/other/wm_menop.htm). of ageing and of estrogen replacement therapyÐespecially oral
estrogenÐsigni®cantly reduces androgen bioavailability after
menopause. The combined effects of reduced production with
Future therapeutic perspectives ageing and the pharmacological effects of oral estrogen dramat-
ically reduce the endogenous and bioavailable androgen milieu
The recognition during the mid-1980s of selective estrogen (Simon, 2002).
receptor modulation provided a unique opportunity to develop The matter of androgen therapy and bone mineral density has
multifunctional drugs. Tamoxifen, the ®rst SERM, is the ®rst been analysed recently (Notelovitz, 2002a). A number of clinical
antiestrogen to be tested successfully for the prevention of breast trials, including appropriately designed randomized blinded stud-
cancer in high-risk women. However, the recognition that SERMs ies, have shown that androgen therapy (testosterone by subcuta-
maintain bone density and lower circulating cholesterol suggested neous pellet implantation or oral methyltestosterone) when
that the prevention of osteoporosis and coronary heart disease combined with estrogen therapy, has an additive effect on bone
would be bene®cial side effects of tamoxifen treatment. This mineral density compared with estrogen-only therapy. These
hypothesis was not pursued in any clinical trial, but an alternate studies were preceded by the earlier empirical use of combination
hypothesisÐthat SERMs could be developed to prevent osteo- estrogen and androgen therapy in clinical practice, and by the
porosis and potentially reduce the risk of breast cancerÐhas been observation in individual women of an anabolic effect on bone of
pursued with raloxifene (Jordan, 2001; Jordan et al., 2001). Recent combined estrogen and androgen treatment. Interestingly, using

evidence from randomized trials suggests that raloxifene prevents clinically available bone marker tests, the rationale for knowing
bone loss and reduces the risk of vertebral fractures (Cranney et al., when to use estrogen and androgen therapy in preference to
2002). The increase in bone density with raloxifene is smaller than estrogen-only when treating women with osteoporosis has been
that seen with other anti-remodelling therapies, such as HRT and proposed (Raisz et al., 1996). Non-response of bone mineral
bisphosphonates (Cranney et al., 2002). The relatively large effect density to adequate estrogen therapy may be indicative of low-
of the drug on vertebral fractures may suggest that raloxifene has a turnover osteoporosis, mainly when appropriate serum levels of
positive effect on other aspects of bone, such as bone quality. On estrogen have been achieved and the urinary excretion of the
the other hand, the very small effect on non-vertebral fractures collagen cross-link peptides are appropriately suppressed. Under
does not support such an effect. these circumstances the anabolic bone-building potential of
There is intense interest in understanding the molecular androgens should be considered (Notelovitz, 2002a). Long-term
mechanisms of action of SERMs at target sites in a woman's clinical trials are needed to con®rm the clinical ef®cacy of
body. Current molecule modelling of the SERM±estrogen receptor estrogen/androgen therapy in osteoporosis prevention and treat-
complex has identi®ed the reason for the promiscuous estrogen- ment. Safety concerns include potential virilizing effects, signi®-
like actions of tamoxifen compared with raloxifene. Future studies cant hepatic events, the generation of an adverse lipoprotein
of the signal transduction pathways of the estrogen receptor alpha- pro®le, and an increased risk of breast cancer (Slayden, 1998;
and beta-SERM complexes hold the promise of new drug Dorgan et al., 2001; Liao and Dickson, 2002). This notwithstand-
discoveries (Jordan 2001; Jordan et al., 2001). An understanding ing, while oral androgens lead to unfavourable changes in
of the targeted actions of this novel drug group will potentially circulating lipid concentrations, there is no evidence that
result in the introduction of new multifunctional medicines with parenteral testosterone administration is likely to have any adverse
applications as preventive agents or treatments of breast cancer, cardiovascular consequences (Simon, 2001; Burger, 2002b). On
osteoporosis, coronary heart disease and endometrial cancer the other hand, results from clinical and non-human primate
(Jordan et al., 2001). studies have shown that androgen administration may induce
Androgens are known to be essential for men's sexual down-regulation of mammary epithelial proliferation and estrogen
performance and well-being, but their contribution to female receptor expression, suggesting that estrogen/androgen therapy
health and sexuality is less well understood. The concept of might reduce the risk of breast cancer associated with estrogen
androgen insuf®ciency in women has been controversial but is replacement therapy (Simon, 2001; Dimitrakakis et al., 2002).
now viewed as potentially having a major impact on the quality Another new perspective in treatment of osteoporosis has been
and well-being of women's lives. There is, at present, a growing proposed by Manolagas' group (Manolagas et al., 2002). Besides
interest in the subject, and a wide spectrum of symptoms and the critical role of sex steroid-controlled osteoblast and osteocyte
signsÐincluding bone loss and muscle wasting that may be apoptosis for the pathogenesis of osteoporosis caused by loss of
associated with androgen insuf®ciency in post-menopausal sex steroids as discussed above, several additional studies from
womenÐhas been strongly identi®ed at a recent consensus that group (Manolagas et al., 2002) have demonstrated that
conference (Bachmann, 2002; Bachmann et al., 2002). Due to osteoblast and osteocyte apoptosis are key pathogenetic mechan-
the potential adverse effects of diminished androgen, the issue of isms in other forms of osteoporosis. Moreover, control of
pharmacological intervention with androgen replacement assumes osteoblast and osteocyte apoptosis would explain the anti-
signi®cance in the female population with androgen insuf®ciency, osteoporotic ef®cacy of the most commonly used drugs. As
especially in the post-menopausal age group where it has been mentioned above, the effect of estrogens (and androgens for this
emphasized that we are only at the beginning of recognizing the matter) is mediated by a novel paradigm of sex steroid action that
importance of androgens (Bachmann, 2002; Burger, 2002b; Sarrel, requires only the ligand-binding domain of the receptor, extra-
2002). Although no precipitous drop in androgen production is nuclear localization of the protein, and sex non-speci®city
apparent at the time of the menopause, androgens decrease slowly (Kousteni et al., 2001). This fact, together with the demonstration
218
of the principle that increased work output of a cell population by Albright, F., Bloomberg, F. and Smith, P.H. (1940) Postmenopausal
osteoporosis. Trans. Assoc. Am. Phys., 55, 298±305.
suppressing apoptosis can augment tissue mass, points to an American College of Physicians (1992) Guidelines for counseling
entirely new approach for the treatment of osteoporosis: one that postmenopausal women about preventive hormone therapy. Ann. Intern.
could lead to cure rather than prevention or slowing of the disease Med., 117, 1038±1041.
process. Studies with the use of different anabolic agents such as Bachmann, G.A. (2002) The hypoandrogenic woman: pathophysiologic
overview. Fertil. Steril., 77 (Suppl. 4), S72±S76.
¯uoride, growth hormone, insulin-like growth factor I, the statins, Bachmann, G.A., Bancroft, J., Braunstein, G., Burger, H., Davis, S.,
and mainly the intermittent administration of parathyroid hormone Dennerstein, L., Godstein, I., Guay, A., Leiblum, S., Lobo, R. et al.
(Neer et al., 2001; Rosen and Bilezikian, 2001) strongly support (2002) Female androgen insuf®ciency: the Princeton consensus statement
on de®nition, classi®cation and assessment. Fertil. Steril., 77, 660±665.
this contention by showing that the anabolic property of
Bassey, E.J. (1995) Exercise in primary prevention of osteoporosis in women.
parathyroid hormone restores bone mineral density in the normal Ann. Rheum. Dis., 54, 861±862.
range and prevents bone fractures to levels much greater than those Bilezikian, J.P. and Silverberg, S.J. (1992) Osteoporosis: a practical approach
seen with anti-resorptive agents (Neer et al., 2001). to the perimenopausal woman. J. Women's Res., 1, 21±27.
Breuil, V. and Euller-Ziegler, L. (2001) Gonadal dysgenesis and bone
The optimal therapeutic modality for osteoporosis, mainly in metabolism. Joint Bone Spine, 68, 26±33.
patients who have already suffered signi®cant bone loss, is clearly Burger, H.G. (2002a) Androgen production in women. Fertil. Steril., 77
an anabolic agent that can restore bone mass by rebuilding bone (Suppl. 4), S3±S5.
within a short period of time. Since daily parathyroid hormone Burger, H.G. (2002b) The role of androgen therapy. Best Pract. Res. Clin.
Obstet. Gynecol., 16, 383±393.
administration increases bone mass by preventing osteoblast Burkman, R.T., Collins, J.A. and Greene, R.A. (2001) Current perspectives on
apoptosis without slowing remodelling, Manolagas' group rea- bene®ts and risks of hormone replacement therapy. Am. J. Obstet.

soned that estrogen receptor ligands with anti-apoptotic but not Gynecol., 185 (Suppl. 2), S13±S23.
anti-resorptive/anti-remodelling properties will expand the pool of Bush, T.L. (1992) Feminine forever revisited: menopausal hormone therapy in
the 1990s. J. Women's Health, 1, 1±14.
mature osteoblasts at sites of new bone formation and allow these Castelo-Branco, C., Vicente, J.J., Figueras, F., Sanjuan, A., Martinez de
cells more time to make bone, to a much greater degree than the Osaba, M.J., Casals, E., Pons, F., Balasch, J. and Vanrell, J.A. (2000)
anti-resorptive agents that also slow remodelling (Manolagas, Comparative effects of estrogens plus androgens and tibolone on bone,
lipid pattern and sexuality in postmenopausal women. Maturitas, 34,
2000; Manolagas et al., 2002). Based on their ®ndings that the 161±168.
non-genotrophic estrogen-like activity of the estrogen receptor can Castelo-Branco, C., GoÂmez, O., Pons, F., Martinez de Osaba, M.J., Balasch, J.
be dissociated from its transcriptional activity with synthetic and Vanrell, J.A. (2003) Secreting ovarian tumors may protect women
ligands (Kousteni et al., 2001), that group of researchers coined the from osteoporosis. Gynecol. Oncol., 88, 149±152.
Cauley, J.A., Seeley, D.G., Ensrud, K., Ettinger, B., Black, D. and Cummings,
acronym ÀNGELS' for estrogen receptor ligands which function S.R. (1995) Study of Osteoporotic Fractures Research Group. Estrogen
as Àctivators of Non-Genotropic Estrogen-Like Signaling' replacement therapy and fractures in older women. Ann. Intern. Med.,
(Manolagas et al., 2002). These ligands lack, either completely 122, 9±16.
or partially, the ability to induce the transcriptional activity of the Cauley, J.A., Black, D.M., Barrett-Connor, E., Harris, F., Shields, K.,
Applegate, W. and Cummings, S.R. (2001) Effects of hormone
estrogen receptor. Evoking the non-genotrophic but not the replacement therapy on clinical fractures and height loss: the Heart and
genotrophic signal, these synthetic ligands may be bone anabolic Estrogen/progestin Replacement Study (HERS). Am. J. Med., 110,
agents, as opposed to natural estrogens or SERMs that are anti- 442±450.
Christiansen, C. (1994) Hormone replacement therapy ± standardized or
resorptive agents. The same ligands may also circumvent the side individually adapted doses?Ðeffect on bone mass. In: Berg, G. and
effects associated with conventional HRT (Manolagas et al., Hammar, M. (eds), The Modern Management of the Menopause. The
2002). Parthenon Publishing Group, New York, pp. 337±341.
In supporting these postulates, the same authors (Manolagas Collaborative Group on Hormonal Factors in Breast Cancer (1997) Breast
cancer and hormone replacement therapy: collaborative reanalysis of data
et al., 2002) have obtained evidence that synthetic ligands with from 51 epidemiological studies of 52 705 women with breast cancer and
potent anti-apoptotic but not genotrophic activity increase bone 108 411 women without breast cancer. Lancet, 350, 1047±1059.
mineral density and bone strength signi®cantly more than Compston, J.E. (1990) Osteoporosis. Clin. Endocrinol., 33, 653±682.
estrogens, in estrogen-de®cient mice, without affecting the uterus Compston, J.E. (1994) Hormone replacement therapy for osteoporosis: clinical
and pathophysiological aspects. Reprod. Endocr. Rev., 3, 209±224.
or the breast. Future studies will be needed to establish the validity Cooper, C., Hannaford, P., Croft, P. and Kay, C.R. (1993) Oral contraceptive
of the ANGELS concept on non-reproductive tissues other than pill use and fractures in women: a prospective study. Bone, 14, 41±45.
bone. However, it is suggested that if non-genotrophic effects of Cooper, G.S. and Thorp, J.M. (1999) FSH levels in relation to hysterectomy
and to unilateral oophorectomy. Obstet. Gynecol., 94, 969±972.
estrogen are as important in other non-reproductive tissues as the
Cranney, A., Tugwell, P., Zytaruk, N., Robinson, V., Weaver, B., Adachi, J.,
evidence suggests, ANGELS may also retain at least some of the Wells, G., Shea, B., Guyatt, G., The Osteoporosis Methodology Group
bene®cial effects of estrogens on the vasomotor, cardiovascular et al. (2002) Meta-analysis of raloxifene for the prevention and treatment
and central nervous systems (Manolagas et al., 2002). of postmenopausal osteoporosis. Endocr. Rev., 23, 524±528.
Cromer, B.A. (1999) Effects of hormonal contraceptives on bone mineral
density. Drug Safety, 20, 213±222.
References Crosignani, P.G. and Vegetti, W. (1996) A practical guide to the diagnosis and
management of amenorrhea. Drugs, 52, 671±681.
Abdalla, H., Hart, D.M., Lindsay, R., Leggate, I. and Hooke, A. (1985) Cummings, S.R., Black, D.M., Thompson, D.E., Applegate, W.B., Barret-
Prevention of bone mineral loss in postmenopausal women by Connor, E., Musliner, T.A., Palermo, L., Prineas, R., Rubin, S.M., Scott,
norethisterone. Obstet. Gynecol., 66, 789±792. J.C. et al. (1998) Effect of alendronate on risk of fracture in women with
Albright, F. (1947) The effect of hormones on osteogenesis in man. Rec. Prog. low bone density but without vertebral fractures: results from the Fracture
Horm. Res., 1, 293±353. Intervention Trial. JAMA, 280, 2077±2082.
Albright, F. and Reifenstein, E.C., Jr (1947) Metabolic bone disease: Cundy, T., Evans, M., Roberts, H., Wattie, D., Ames, R. and Reid, I.R. (1991)
osteoporosis. In: Albright, F. and Reifenstein, E.C. (eds), The Bone density in women receiving depot medroxyprogesterone acetate for
Parathyroid Glands and Metabolic Bone Disease. Baltimore: Williams contraception. Br. Med. J., 303, 13±16.
& Wilkins, pp. 145±204. Cundy, T., Farquhar, C.M., Cornish, J. and Reid, I.R. (1996) Short-term
219
J.Balasch
effects of high dose oral medroxyprogesterone acetate on bone density in (1997) Effects on bone mineral density of low-dose oral contraceptives
premenopausal women. J. Clin. Endocrinol. Metab., 81, 1014±1017. compared to and combined with physical activity. Contraception, 55,
Dawood, M.Y. (1994) Hormonal therapies for endometriosis: implications for 87±90.
bone metabolism. Acta Obstet. Gynecol. Scand., 159, 22±34. Heaney, R.P. (1987) The role of nutrition in prevention and management of
Dayal, M. and Barnhart, K.T. (2001) Noncontraceptive bene®ts and osteoporosis. Clin. Obstet. Gynecol., 30, 833±846.
therapeutic uses of the oral contraceptive pill. Semin. Reprod. Med., 19, Hemminki, E. and Sihvo, S. (1993) A review of postmenopausal hormone
295±303. therapy recommendations: potential for selection bias. Obstet. Gynecol.,
Deary, A.J., Breeze, A.C.G., Barlow, D. and Prentice, A. (2002) 82, 1021±1028.
Gonadotrophin-releasing hormone analogues for endometriosis: bone Herrington, D.M. (1999) The HERS trial results: paradigms lost? Ann. Intern.
mineral density (Cochrane Review). In: The Cochrane Library, Issue 1, Med., 131, 463±466.
2002. Hornstein, M.D. and Barbieri, R.L. (1988) Endocrine therapy of
DeCherney, A. (1993) Physiologic and pharmacologic effects of estrogen and endometriosis: Danazol and the synthetic progestins. In: Barbieri, R.L.
progestins on bone. J. Reprod. Med., 38, (Suppl.), 1007±1014. and Schiff, I. (eds), Reproductive Endocrine Therapeutics. Alan R. Liss,
Delmas, P.D. (1995) Biochemical markers for the assessment of bone Inc., New York, pp. 155±185.
turnover. In: Riggs, B.L. and Melton, L.J., III (eds), Osteoporosis: Horowitz, M., Wishart, J.M., Need, A.G., Morris, H.A. and Nordin, B.E.
Etiology, Diagnosis, and Management, 2nd edition. Lippincott-Raven (1993) Effects of norethisterone on bone related biochemical variables
Publishers, Philadelphia, pp. 319±333. and forearm bone mineral in post-menopausal osteoporosis. Clin.
DeSouza, M.J., Miller, B.E., Sequenzia, L.C., Luciano, A.A., Ulreich, S., Endocrinol., 39, 649±655.
Stier, S., Prestwood, K. and Lesley, B.L. (1997) Bone health is not Hreshchyshyn, M.M., Hopkins, A., Zylstra, S. and Anbar, M. (1988) Effects of
affected by luteal phase abnormalities and decreased ovarian progesterone natural menopause, hysterectomy, and oophorectomy on lumbar spine
production in female runners. J. Clin. Endocrinol. Metab., 82, and femoral neck bone densities. Obstet. Gynecol., 72, 631±638.
2867±2876. Hulka, B.S. (1994) Links between hormone replacement therapy and
Dimitrakakis, C., Zhou, J. and Bondy, C.A. (2002) Androgens and mammary neoplasia. Fertil. Steril., 62 (Suppl. 2), 168S±175S.

growth and neoplasia. Fertil. Steril., 77, (Suppl. 4), S26±S33. Hulka, B.S. and Stark, A.T. (1995) Breast cancer: cause and prevention.
Dixon, J.M. (2001) Hormone replacement therapy and the breast: we should Lancet, 346, 883±887.
worry about the increase in the risk of breast cancer. Br. Med. J., 323, Hulley, S., Furberg, C., Barret-Connor, E., Cauley, J., Grady, D., Haskell, W.,
1381±1382. Knopp, R., Lowery, M., Satter®eld, S., Schrott, H., Vittinghoff, E.,
Dorgan, J.F., Boudou, P., Stanczyk, F.Z., Longcope, C., Tejpar, E.A., Falk, Hunninghake, D. for the HERS Research Group (2002)
R.T., Schussler, N. and Stephenson, H.E., Jr (2001) Sources of elevated Noncardiovascular disease outcomes during 6.8 years of hormone
serum androgens in postmenopausal women who develop breast cancer. therapy. Heart and Estrogen/progestin Replacement Study follow-up
Cancer Epidemiol. Biomarkers Prev., 10, 407±410. (HERS II). JAMA, 288, 58±66.
Drinkwater, B.L., Nilson, K., Ott, S. and Chesnut, C.H., III (1986) Bone HumphreyL.L., Chan, B.K.S. and Sox, H.C. (2002) Postmenopausal hormone
mineral density after resumption of menses in amenorrheic athletes. replacement therapy and the primary prevention of cardiovascular
JAMA, 256, 380±382.
disease. Ann. Intern. Med., 137, 273±284.
Edmonds, D.K. (1996) Add-back therapy in the treatment of endometriosis:
Ishida, Y. and Heersche, J.N.M. (1999) Progesterone- and dexamethasone-
the European experience. Br. J. Obstet. Gynecol., 103 (Suppl. 14), 10±13.
dependent osteoprogenitors in bone cell populations derived from rat
Eldred, J.M., Haynes, P.J. and Thomas, E.J. (1992) A randomized double-
vertebrae are different and distinct. Endocrinology, 140, 3210±3218.
blind placebo controlled trial of the effects on bone metabolism of the
Ishida,Y. and Heersche, J.N.M. (2002) Pharmacologic doses of
combination of nafarelin acetate and norethisterone. Clin. Endocrinol.
medroxyprogesterone may cause bone loss through glucocorticoid
(Oxf.), 37, 354±359.
activity: an hypothesis. Osteoporos. Int., 13, 601±605.
Ensrud, K.E., Palermo, L., Black, D.M., Cauley, J., Jergas, M., Orwoll, E.S.,
Jeng, M.H., Parker, J. and Jordan, C. (1992) Estrogenic potential of progestins
Nevitt, M.C., Fox, K.M. and Cummings, S.R. (1995) Hip and calcaneal
bone loss increase with advancing age: longitudinal results from the Study in oral contraceptives to stimulate human breast cancer cell proliferation.
of Osteoporotic Fractures. J. Bone Miner. Res., 10, 1778±1787. Cancer Res., 52, 6539±6546.
Eriksen, E.F., Colvard, D.S., Berg, N.J., Graham, M.L., Mann, K.G., Johnston, C.C. and Melton, L.J., III (1995) Bone densitometry. In: Riggs, B.L.
Spelsberg, T.C. and Riggs, B.L. (1988) Evidence of estrogen receptors and Melton, L.J., III (eds), Osteoporosis: Etiology, Diagnosis, and
in normal human osteoblast-like cells. Science, 241, 84±86. Management, 2nd edition. Lippincott-Raven Publishers, Philadelphia, pp.
Felson, D.T., Zhang, Y., Hanna, M.T., Kiel, D.P., Wilson, P.W. and Anderson, 275±297.
J.J. (1993) The effect of postmenopausal estrogen therapy on bone density Jordan, V.C. (2001) The past, present, and future of selective estrogen receptor
in elderly women. N. Engl. J. Med., 329, 1141±1146. modulation. Ann. N. Y. Acad. Sci., 949, 72±79.
Friedrich, M.J. (2001) Recent studies bring risks, bene®ts of hormone Jordan, V.C., Jeng, M.H., Catherino, W.H. and Parker, C.J. (1993) The
replacement therapy under scrutiny. J. Natl Cancer Inst., 93, 1287±1288. estrogenic activity of synthetic progestin used in oral contraceptives.
Fujimura, R., Ashizawa, N., Watanabe, M., Mukai, N., Amagai, H., Cancer, 71, 1501±1505.
Fukubayashi, T., Hayashi, K., Tokuyama, K. and Suzuki, M. (1997) Jordan, V.C., Gapstur, S. and Morrow, M. (2001) Selective estrogen receptor
Effect of resistance exercise training on bone formation and resorption in modulation and reduction in risk of breast cancer, osteoporosis, and
young male subjects assessed by biomarker of bone metabolism. J. Bone coronary heart disease. J. Natl Cancer. Inst., 93, 1449±1457.
Miner. Res. 12, 656±662. Kaunitz, A.M. (1998) Injectable depot medroxyprogesterone acetate
Gallagher, J.C., Kabel, W.T. and Goldgar, D. (1991) Effect of progestin contraception: an update for U.S. clinicians. Int. J. Fertil. Womens
therapy on cortical and trabecular bone: comparison with estrogen. Am. J. Med., 43, 73±83.
Med., 90, 171±178. Kaunitz, A.M. (2002) Current concepts regarding use of DMPA. J. Reprod.
Grady, D. and Cummings, S.R. (2001) Postmenopausal hormone therapy for Med., 47, (Suppl. 9), 785±789.
prevention of fractures: how good is the evidence? JAMA, 285, Khosla, S.K., Riggs, B.L. and Melton, L.J., III (1995) Clinical spectrum. In:
2909±2910. Riggs, B.L. and Melton, L.J., III (eds), Osteoporosis: Etiology, Diagnosis,
Grady, D., Rubin, S.M., Petiti, D.B., Fox, C.S., Black, D., Ettinger, B., Ernster, and Management, 2nd edition. Lippincott-Raven Publishers, Philadelphia,
V.L. and Cummings, S.R. (1992) Hormone therapy to prevent disease and pp. 205±223.
prolong life in postmenopausal women. Ann. Intern. Med., 117, Kiesel, L., Schweppe, K.W., Sillem, M. and SiebzehnruÈbl, E. (1996) Should
1016±1037. add-back therapy for endometriosis be deferred for optimal results? Br. J.
Gregoriou, O., Kouskouni, E., Bakas, P., Konidaris, S., Papadias, K., Obstet. Gynecol., 103 (Suppl. 14), 15±17.
Kalovidouris, A. and Creatsas, G. (2000) Bone mineral density in Komm, B.S., Terpening, C.M., Benz, N.J., Graeme, K.A., Gallegos, A., Korc,
women with idiopathic hirsutism. Gynecol. Endocrinol., 14, 364±368. M., Greene, G.L., O'Malley, B.W. and Haussler, M.R. (1988) Estrogen
Hankinson, S.E., Hunter, D.J., Colditz, G.A., Willett, W.C., Stampfer, M.J., binding, receptor mRNA, and biologic response in osteoblast-like
Rosner, B., Hennekens, C.H. and Speizer, F.E. (1993) Tubal ligation, osteosarcoma cells. Science, 241, 81±84.
hysterectomy, and risk of ovarian cancer. A prospective study. JAMA, Kousteni, S., Bellido, T., Plotkin, L.I., O'Brien, C.A., Bodenner, D.L., Han, L.,
270, 2813±2818. Han, K., DiGregorio, G.B., Katzenellenbogen, J.A., Katzenellenbogen,
Hartard, M., Bottermann, P., Bartenstein, P., Jeschke, D. and Schwaiger, M. B.S. et al. (2001) Nongenotropic, sex-nonspeci®c signaling through the
220
estrogen or androgen receptors: dissociation from transcriptional activity. Notelovitz, M. (2002a) Androgen effects on bone and muscle. Fertil. Steril.,
Cell, 104, 719±730. 77 (Suppl. 4), S34±S41.
Lacey, J.V., Jr, Mink, P.J., Lubin, J.H., Sherman, M.E., Troisi, R., Hartge, P., Notelovitz, M. (2002b) Overview of bone mineral density in postmenopausal
Schtzkin, A. and Schrairer, C. (2002) Menopausal hormone replacement women. J. Reprod. Med., 47 (Suppl. 1), 71±81.
therapy and risk of ovarian cancer. JAMA, 288, 334±341. Orwoll, E.S. and Nelson, H.D. (1999) Does estrogen adequately protect
Laine, C. (2002) Postmenopausal hormone replacement therapy: how could postmenopausal women against osteoporosis: an iconoclastic perspective.
we have been so wrong? Ann. Intern. Med., 137, 290. J. Clin. Endocrinol. Metab., 84, 1872±1874.
Liao, D.J. and Dickson, R.B. (2002) Roles of androgens in the development, Orwoll, E.S., Yuzpe, A.A., Burry, K.A., Heinrichs, L., Buttram, V.C. and
growth, and carcinogenesis of the mammary gland. J. Steroid Biochem. Hornstein, M.D. (1994) Nafarelin therapy in endometriosis: long-term
Mol. Biol., 80, 175±189. effects on bone mineral density. Am. J. Obstet. Gynecol., 171, 1221±1225.
Lindsay, R. (1987) The menopause: sex steroids and osteoporosis. Clin. Paoletti, A.M., Serra, G.G., Cagnacci, A., Vacca, A.M., Guerreiro, S., Solla, E.
Obstet. Gynecol., 30, 847±859. and Melis, G.B. (1996) Spontaneous reversibility of bone loss induced by
Lindsay, R. (1995) Estrogen de®ciency. In: Riggs, B.L. and Melton, L.J., III gonadotropin-releasing hormone analogue treatment. Fertil. Steril., 65,
(eds), Osteoporosis: Etiology, Diagnosis, and Management, 2nd edition. 707±710.
Lippincott-Raven Publishers, Philadelphia, pp. 133±160. Par®tt, A.M. (1987) Bone remodeling and bone loss: understanding the
Lobo, R.A. and Speroff, L. (1994) International consensus conference on pathophysiology of osteoporosis. Clin. Obstet. Gynecol., 30, 789±811.
postmenopausal hormone therapy and the cardiovascular system. Fertil. Pike, M.C. and Spicer, D.V. (2000) Hormonal contraception and
Steril., 61, 592±595. chemoprevention of female cancers. Endocr. Relat. Cancer, 7, 73±83.
Luukkainen, T., Pakarinen, P. and Toivonen, J. (2001) Progestin-releasing Pinkerton, J.V. and Santen, R. (2002) Use of alternatives to estrogen for
intrauterine systems. Semin. Reprod. Med., 19, 355±363. treatment of menopause. Minerva Endocrinol., 27, 21±41.
Mais, V., Fruzzetti, F., Aiossa, S., Paoletti, A.M., Guerriero, S. and Melis, Prior, J.C. (1990) Progesterone as a bone-trophic hormone. Endocr. Rev., 11,
G.B. (1993) Bone metabolism in young women taking a monophasic pill 386±398.
containing 20 mg ethinylestradiol. Contraception, 48, 445±452. Prior, J.C., Vigna, Y.M., Barr, S.I., Rexworthy, C. and Lentle, B.C. (1994)

Manolagas, S.C. (1999) Cell number versus cell vigor ± what really matters to Cyclic medroxyprogesterone treatment increases bone density: a
a regenerating skeleton? Endocrinology, 140, 4377±4381. controlled trial in active women with menstrual cycle disturbance. Am.
Manolagas, S.C. (2000) Birth and death of bone cells: basic regulatory J. Med., 96, 521±530.
mechanisms and implications for the pathogenesis and treatment of Raisz, L.G., Wiita, B., Artis, A., Bowen, A., Schwartz, S., Trahiotis, M.,
osteoporosis. Endocr. Rev., 21, 115±137. Shoukri, K. and Smith, J. (1996) Comparison of the effects of estrogen
Manolagas, S.C., Kousteni, S. and Jilka, R.L. (2002) Sex steroids and bone. alone plus and estrogen plus androgen on biochemical markers of bone
Recent Prog. Horm. Res., 57, 385±409. formation and resorption in postmenopausal women. J. Clin. Endocrinol.
Marcus, R. (2002) Post-menopausal osteoporosis. Best Pract. Res. Clin. Metab., 81, 37±43
Obstet. Gynecol., 16, 309±327. Randall, T. (1993) Women need more and better information on menopause
Marcus, R., Cann, C. and Madvig, D. (1985) Menstrual function and bone from their physicians. JAMA, 270, 1664.
mass in elite women distance runners. Ann. Intern. Med., 102, 158±163. Reeve, J. (2000) How do women develop fragile bones? J. Steroid Biochem.
McClung, M.R., Geusens, P. and Miller, P.D. (2001) Effect of risedronate on Mol. Biol., 74, 375±381.
the risk of hip fracture in elderly women: HIP Intervention Program. N. Reginster, J.Y., Bruyere, O., Audran, M., Avouac, B., Body, J.J., Bouvenot,
Engl. J. Med., 344, 333±340. G., Brandi, M.L., Gennari, C., Kaufman, J.M., Lemmel, E.M., Vanhaelst,
McNagny, S.E. and Wenger, N.K. (2002) Postmenopausal hormone- L., Weryha, G., Devogelaer, J.P., On behalf of the Group for the Respect
replacement therapy. N. Engl. J. Med., 346, 63. of Ethics and Excellence in Science (GREES) (2000) Do estrogens
Meckstroth, K.R. and Darney, P.D. (2001) Implant contraception. Semin. effectively prevent osteoporosis-related fractures? Calcif. Tissue Int., 67,
Reprod. Med., 19, 339±354. 191±194.
MichaeÈlsson, K., Baron, J.A., Farahmand, B.Y., Persson, I. and Ljunghall, S. Rekers, H. (1991) Mastering the menopause. In: Burger, H. and Boulet, M.
(1999) Oral contraceptive use and risk of hip fracture: a case-control (eds), A Portrait of the Menopause. The Parthenon Publishing Group,
study. Lancet, 353, 1481±1484. Carnforth, Lancs, UK, pp. 23±43.
Moghissi, K.S. (1996) Add-back therapy in the treatment of endometriosis: the Riggs, B.L. and Khosla, S. (1995) Practical clinical management. In: Riggs,
North American experience. Br. J. Obstet. Gynecol., 103 (Suppl. 14), 14. B.L. and Melton, L.J., III (eds), Osteoporosis: Etiology, Diagnosis, and
Narod, S.A., Sun, P., Ghadirian, P., Lynch, H., Isaacs, C., Garber, J., Weber, Management, 2nd edition. Lippincott-Raven Publishers, Philadelphia, pp.
B., Karlan, B., Fishman, D., Rosen, B. et al. (2001) Tubal ligation and 487±502.
risk of ovarian cancer in carriers of BRCA1 or BRCA2 mutations: a case- Riggs, B.L., Khosla, S. and Melton, J., III (2002) Sex steroids and the
control study. Lancet, 357, 1467±1470. construction and conservation of the adult skeleton. Endocr. Rev., 23,
Neer, R.M., Arnaud, C.D., Zanchetta, J.R., Prince, R., Gaich, G.A., Reginster, 279±302.
J.Y., Hodsman, A.B., Eriksen, E.F., Ish-Shalom, S., Genant, H.K. et al. Riis, B.J., Christiansen, C., Johansen, J.S. and Jacobson, J. (1990) Is it possible
(2001) Effect of parathyroid hormone (1-34) on fractures and bone to prevent bone loss in young women treated with luteinizing hormone-
mineral density in postmenopausal women with osteoporosis. N. Engl. J. releasing hormone agonists? J. Clin. Endocrinol. Metab., 70, 920±924.
Med., 344, 1434±1441. Riman, T., Dickman, P.W., Nilsson, S., Correia, N., Nordlinder, H.,
Nelson, H.D., Rizzo, J., Harris, E., Cauley, J., Ensrud, K., Bauer, D.C., Orwoll, Magnusson, C.M., Weiderpass, E. and Persson, I.R. (2002a) Hormone
E.: Study of Osteoporotic Fractures Research Group (2002a) Osteoporosis replacement therapy and the risk of invasive epithelial ovarian cancer in
and fractures in postmenopausal women using estrogen. Arch. Intern. Swedish women. J. Natl Cancer Inst., 94, 497±504.
Med., 162, 2278±2284. Riman, T., Dickman, P.W., Nilsson, S., Correia, N., Nordlinder, H.,
Nelson, H.D., Humphrey, L.L., Nygren, P., Teutsch, S.M. and Allan, J.D. Magnusson, C.M. and Persson, I.R. (2002b) Risk factors for invasive
(2002b) Postmenopausal hormone replacement therapy. Scienti®c review. epithelial ovarian cancer: results from a Swedish case-control study. Am.
JAMA, 288, 872±881. J. Epidemiol., 156, 363±373.
NIH Consensus Development Panel on Osteoporosis Prevention, Diagnosis, Rodan, G.A. and Rodan, S.B. (1995) The cells of bone. In: Riggs, B.L. and
and Therapy (2001) Osteoporosis prevention, diagnosis, and therapy. Melton, L.J., III (eds), Osteoporosis: Etiology, Diagnosis, and
JAMA, 285, 785±795. Management, 2nd edition. Lippincott-Raven Publishers, Philadelphia,
Noller, K.L. (2002) Estrogen replacement therapy and risk of ovarian cancer. pp. 1±39.
JAMA, 288, 368±369. Rodriguez, C., Patel, A.V., Calle, E.E., Jacob, E.J. and Thun, M.J. (2001)
Nordin, B.E., Need, A.G., Chatterton, B.E., Horowitz, M. and Morris, H.A. Estrogen replacement therapy and ovarian cancer mortality in a large
(1990) The relative contributions of age and years since menopause to prospective study of US women. JAMA, 285, 1460±1465.
postmenopausal bone loss. J. Clin. Endocrinol. Metab., 70, 83±88. Rosen, C.J. and Bilezikian, J.P. (2001) Anabolic therapy for osteoporosis. J.
Notelovitz, M. (1988) The non-hormonal management of menopause. In: Clin. Endocrinol. Metab., 86, 957±964.
Studd, J.W.W. and Whitehead, M.I. (eds), The Menopause. Blackwell Rosen, H.N. and Rosenblatt, M. (2002) Epidemiology and causes of
Scienti®c Publications, Oxford, pp. 102±115. osteoporosis. UpToDate, February 25, 2002.
Notelovitz, M. (1993) Osteoporosis: screening, prevention, and management. Rosenberg, L. (1993) Hormone replacement therapy: the need for
Fertil. Steril., 59, 707±725. reconsideration. Am. J. Public Health, 83, 1670±1673.
221
J.Balasch
Rosenblatt, K.A., Thomas, D.B. and The World Health Organization Speroff, L., Glass, R.H. and Kase, N.G. (1999) Clinical Gynecological
Collaborative Study of Neoplasia and Steroid Contraceptives (1996) Endocrinology and Infertility. 6th edition. Lippincott Williams &
Reduced risk of ovarian cancer in women with a tubal ligation or Wilkins, Baltimore.
hysterectomy. Cancer Epidemiol. Biomarkers Prev., 5, 933±935. Spicer, D.V., Pike, M.C., Pike, A., Rude, R., Shoupe, D. and Richardson, J.
Sarrel, P.M. (2002) Androgen de®ciency: menopause and estrogen-related (1993) Pilot trial of a gonadotropin releasing hormone agonist with
factors. Fertil. Steril., 77 (Suppl. 4), S63±S67. replacement hormones as a prototype contraceptive to prevent breast
Schlechte, J.A., Sherman, B. and Martin, B. (1983) Bone density in cancer. Contraception, 47, 427±444.
amenorrheic women with and without hyperprolactinemia. J. Clin. Stampfer, M.J. and Hankinson, S.E. (1997) Estrogen and breast cancer: the
Endocrinol. Metab., 56, 1120±1123. duration connection. Menopause, 4, 123±124.
Schot, L.P.C. and Schuurs, A.H.W.M. (1990) Sex steroids and osteoporosis: Stevenson, J.C. (1990) Pathogenesis, prevention, and treatment of
effects of de®ciencies and substitutive treatments. J. Steroid Biochem. osteoporosis. Obstet. Gynecol., 75, 36S±41S.
Mol. Biol., 37, 167±182. Studd, J. and Leather, A.T. (1996) The need for add-back with gonadotrophin-
Siddle, N., Sarrel, P. and Whitehead, M. (1987) The effect of hysterectomy on releasing hormone agonist therapy. Br. J. Obstet. Gynecol., 103 (Suppl.
the age at ovarian failure: identi®cation of a subgroup of women with 14), 1±4.
premature loss of ovarian function and literature review. Fertil. Steril., 47, Sugimoto, A.K., Hodsman, A.B. and Nisker, J.A. (1993) Long-term
94±100. gonadotropin-releasing hormone agonist with standard postmenopausal
Simon, J.A. (2001) Safety of estrogen/androgen regimens. J. Reprod. Med., 46 estrogen replacement failed to prevent vertebral bone loss in
(3 Suppl.), 281±290. premenopausal women. Fertil. Steril., 60, 672±674.
Simon, J.A. (2002) Estrogen replacement therapy: effects on the endogenous The Writing Group for the PEPI Trial (1996) Effects of hormone therapy on
androgen milieu. Fertil. Steril., 77 (Suppl. 4), S77±S82. bone mineral density. Results from the post-menopausal estrogen/
Simon, J.A., Hunninghake, D.B., Agarwal, S.K., Lin, F., Cauley, J.A., Ireland, progesterone intervention (PEPI) trial. JAMA, 276, 1389±1396.
C.C. and Pickar, J.H. (2001) Effect of estrogen plus progestin on risk for Tudor-Locke, C. and McColl, R.S. (2000) Factors related to variation in
biliary tract surgery in postmenopausal women with coronary artery premenopausal bone mineral status: a health promotion approach.

disease. The Heart and Estrogen/progestin Replacement Study. Ann. Osteoporos. Int., 11, 1±24.
Intern. Med., 135, 493±501. Vastag, B. (2002) Hormone replacement therapy falls out of favor with expert
Slayden, S.M. (1998) Risks of menopausal androgen supplementation. Semin. committee. JAMA, 287, 1923±1924.
Reprod. Endocrinol., 16, 145±152. Vessey, M., Mant, J. and Painter, R. (1998) Oral contraception and other
Slemenda, C., Longcope, C., Peacock, M., Hui, S. and Johnston, C.C. (1996) factors in relation to hospital referral for fracture. Findings in a large
Sex steroids, bone mass, and bone loss: a prospective study of pre-, peri-, cohort study. Contraception, 57, 231±235.
and post-menopausal women. J. Clin. Invest., 97, 14±21. Waller, K., Reim, J., Fenster, L., Swan, S.H., Brumback, B., Windham, G.C.,
Slootweg, M.C., Ederveen, A.G.H., Schot, L.P.C., Schoonen, W.G.E.J. and Lasley, B., Ettinger, B. and Marcus, R. (1996). Bone mass and subtle
Kloosterboer, H.J. (1992) Oestrogen and progestogen synergistically abnormalities in ovulatory function in healthy women. J. Clin.
stimulate human and rat osteoblast proliferation. J. Endocrinol., 133, Endocrinol. Metab., 81, 663±668.
R5±R8. Wang, L. and Arnold, K. (2002) Hormone replacement therapy may be
Souza, A.Z., Fonseca, A.M., Izzo, V.M., Clauzet, R.M. and Salvatore, C.A. associated with increased ovarian cancer risk. J. Natl Cancer Inst., 94,
(1986) Ovarian histology and function after total abdominal 467.
hysterectomy. Obstet. Gynecol., 68, 847±849. Warren, M.P., Brooks-Gunn, J., Hamilton, L.H., Warrent, L.J. and Hamilton,
Sowers, M.F.R. and Galuska, D.A. (1993) Epidemiology of bone mass in W.G. (1991) Lack of bone accretion and amenorrhea: evidence for a
premenopausal women. Epidemiol. Rev., 15, 374±398. relative osteopenia in weight-bearing bones. J. Clin. Endocrinol. Metab.,
Sowers, M., Clark, M.K., Hollis, B., Wallace, R.B. and Jannausch, M. (1992) 72, 847±853.
Radial bone mineral density in pre- and perimenopausal women: a Wei, L.L., Leach, M.N., Miner, R.S. and Demers, L.M. (1993) Evidence for
prospective study of rates and risk factors of loss. J. Bone Miner. Res., 7, progesterone receptors in human osteoblast-like cells. Biochem. Biophys.
647±657. Res. Commun., 195, 525±532.
Sowers, M., Randolph, J.F., Crutch®eld, M., Jannausch, M.L., Shapiro, B., Writing Group for the Women's Health Initiative Investigators (2002) Risks
Zhang, B. and LaPietra, M. (1998) Urinary ovarian and gonadotropin and bene®ts of estrogen plus progestin in healthy postmenopausal women.
hormone levels in premenopausal women with low bone mass. J. Bone Principal results from the Women's Health Initiative randomized
Miner. Res., 13, 1191±1202. controlled trial. JAMA, 288, 321±333.
Speroff, L., Rowan, J., Symons, J., Genant, H. and Wilborn, W. (1996) The Zborowski, J.V., Cauley, J.A., Talbott, E.O., Guzick, D.S. and Winters, S.J.
comparative effect on bone density, endometrium and lipids of (2000) Bone mineral density, androgens and the polycystic ovary: the
continuous hormones as replacement therapy (Chart Study). JAMA, complex and controversial issue of androgenic in¯uence in female bone.
276, 1397±1403. J. Clin. Endocrinol. Metab., 67, 937±943.
222

Sex Steroids and Bone: Current Perspectives: Juan Balasch

Uploaded by

Copyright:

Available Formats

Sex Steroids and Bone: Current Perspectives: Juan Balasch

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sex Steroids and Bone: Current Perspectives: Juan Balasch

Uploaded by

Copyright:

Available Formats

dmg017

Sex steroids and bone: current perspectives

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Key words: bone homeostasis/bone remodelling/cytokines/hormone replacement therapy/osteoporosis/ovarian steroids

Introduction osteoblast function, and this is still considered today to be a

ã European Society of Human Reproduction and Embryology 207

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

carriers of BRCA1 and BRCA2 mutations (Hankinson et al., Scoliosis

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Figure 1. Schematic representation of the four major interactive factors

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

Downloaded from humupd.oxfordjournals.org by guest on January 12, 2011

You might also like