• weight;
  • body fat mass;
  • lean body mass;
  • estrogen;
  • menopause


  1. Top of page
  2. Abstract
  7. References

The aim of this study was to study the influence of hormone replacement therapy (HRT) on weight changes, body composition, and bone mass in early postmenopausal women in a partly randomized comprehensive cohort study design. A total of 2016 women ages 45–58 years from 3 months to 2 years past last menstrual bleeding were included. One thousand were randomly assigned to HRT or no HRT in an open trial, whereas the others were allocated according to their preferences. All were followed for 5 years for body weight, bone mass, and body composition measurements. Body weight increased less over the 5 years in women randomized to HRT (1.94 ± 4.86 kg) than in women randomized to no HRT (2.57 ± 4.63, p = 0.046). A similar pattern was seen in the group receiving HRT or not by their own choice. The smaller weight gain in women on HRT was almost entirely caused by a lesser gain in fat. The main determinant of the weight gain was a decline in physical fitness. Women opting for HRT had a significantly lower body weight at inclusion than the other participants, but the results in the self-selected part of the study followed the pattern found in the randomized part. The change in fat mass was the strongest predictor of bone changes in untreated women, whereas the change in lean body mass was the strongest predictor when HRT was given. Body weight increases after the menopause. The gain in weight is related to a decrease in working capacity. HRT is associated with a smaller increase in fat mass after menopause. Fat gain protects against bone loss in untreated women but not in HRT-treated women. The data suggest that women's attitudes to HRT are more positive if they have low body weight, but there is no evidence that the conclusions in this study are skewed by selection bias.


  1. Top of page
  2. Abstract
  7. References

POSTMENOPAUSAL HORMONE REPLACEMENT therapy (HRT) is a controversial treatment with a wide spectrum of long-term benefits and side effects. In most women, the decision to start HRT is based on present menopausal symptoms such as hot flashes, sleeping disturbances, or urogenital discomfort.(1) The positive effect of HRT on bone mineral density (BMD)(2–5) and fracture rates(6–9) has been shown in many studies. Furthermore, HRT seems to protect against loss of lean body mass (LBM), which is seen during the first 2 years after menopause.(10) A steep decline in BMD at the time of menopause has been established, and some women experience an increase in body weight during the same period of life.(11, 12) Many women believe that HRT aggravates the increase in body weight and refuse the treatment for that reason.(12) Up to 18% of HRT users report weight gain as the main reason for stopping treatment.(13) However, this view was not supported by a recent review from the Cochrane Database,(14) which reported no difference in weight gain between women receiving and women not receiving HRT (n = 1497 combined from seven trials).(14, 15) Furthermore, a more pronounced weight loss has been seen in obese women on HRT than in obese women not receiving HRT.(16) It has also been observed that most women gain weight during perimenopause irrespective of HRT use or not; the weight gain is more strongly related to lack of exercise and alcohol consumption.(17) The changes in weight after menopause may thus be influenced by many factors besides estrogen(18, 19) and may to some extent be linked to menopause itself.(17) Available studies have not allowed definite conclusions on changes in body composition (the distribution between fat mass and LBM) during menopause and the influence of HRT on body composition.(14)

Cross-sectional studies have shown that a high body weight predicts a high BMD. Furthermore, clinical studies on obese subjects have demonstrated that voluntary weight loss is followed by a corresponding loss of bone mineral.(20, 21) However, the studies, which have the advantage of studying bone during large weight changes, are limited by a short follow-up time of less than 1 year. The nature of the substantial bone changes in obese subjects during weight reduction is not yet established. Reduced extraovarial estrogen synthesis caused by reduced fat mass (FM) has been suggested, but not shown, and newer studies point in the direction that secondary hyperparathyroidism may play a significant role, at least in obese subjects, on a weight-reducing diet.(22) As osteoporotic fractures are more frequent in lean women, and obesity seems to have a protective effect, the interactions between weight, BMD, and HRT are of significant epidemiological and clinical interest.

The aim of this longitudinal study was to present corresponding data on changes in weight, body composition, and bone mineral over a period of 5 years after menopause. Furthermore, the effect on weight changes and body composition of HRT received after randomization and by own choice is evaluated.


  1. Top of page
  2. Abstract
  7. References

The Danish Osteoporosis Prevention Study (DOPS) is an ongoing long-term comprehensive cohort multicenter study of osteoporotic fracture prevention in postmenopausal women through the use of HRT. The study was approved by the regional ethics committee in 1990 (1990/1821). The participants were recruited by a short questionnaire concerning menopausal status mailed to a random sample of women (n = 47,720), 45–58 years old. Women returning the questionnaires, willing to participate, and fulfilling the inclusion criteria were invited to receive further information and clinical examination. Inclusion criteria were as follows: (1) women with intact uterus aged 45–58 years and 3–24 months past last menstrual bleeding or experiencing perimenopausal symptoms (including irregular menstruation)—the latter combined with elevated serum follicular stimulating hormone (FSH), and (2) hysterectomized women aged 45–52 years with elevated FSH. Exclusion criteria were as follows: (1) metabolic bone disease, including osteoporosis defined as nontraumatic vertebral fractures on X-ray; (2) current estrogen use or estrogen use within the past 3 months; (3) current or past treatment with glucocorticoids >6 months; (4) current or past malignancy; (5) newly diagnosed or uncontrolled chronic disease; or (6) alcohol or drug addiction.

After written informed consent (Helsinki II) 2016 women were enrolled and allocated to four treatment groups between November 1990 and March 1993. These four groups were as follows: (1) randomized to HRT (see below for specific drugs), (2) randomized to no HRT, (3) HRT by own choice, and (4) no HRT by own choice. Participants were block-randomized in groups of 10 by the envelope method. The trial groups were not blinded and HRT was provided free of charge. For details on the design and inclusion see Mosekilde et al.(23) The independent Good Clinical Practice (GCP) unit at Aarhus University Hospital monitored the study.

Study drugs

First-line HRT was (1) sequential oral estrogen and progestogen (28 day cycle: one tablet containing 2 mg estradiol for the first 12 days, one tablet containing 2 mg estradiol and 1 mg norethisterone acetate for the next 10 days, and one tablet containing 1 mg estradiol for 6 days; Trisequens, Novo Nordisk, Bagsværd, Denmark) in women with intact uterus (n = 407) and (2) oral continuous estradiol 2 mg/day (Estrofem; Novo Nordisk) in hysterectomized women (n = 95). If a change of HRT type for reasons not requiring permanent discontinuation was requested and the woman accepted, a number of alternatives were available.(24)

Evaluation of weight, body composition, and bone mineral content

At inclusion and after 6 months, 1 year, 2 years, and 5 years, the women had their body weight measured in light indoor clothing without shoes using a calibrated scale (Detecto electronic scale; Detecto Scale Co., St. Louis, MO, USA).

Height was measured without shoes using a calibrated scale (Stadiometer; SECA, Vogd & Halke, Hamburg, Germany).

Body composition and regional bone mineral were measured at inclusion and after 1, 2, and 5 years using Hologic QDR whole-body scanners (Hologic Inc., Waltham, MA, USA). FM (kg), LBM (kg), and total body bone mineral (TBBM, kg; in vivo precision errors, 1.1%, 1.6%, and 1.6%, respectively) were measured by whole body scans in the pencil beam mode. The regional measurements included bone mineral content (BMC, g) and BMD (g/cm2) of the right hip, the lumbar spine (L2-L4), and the ultradistal forearm (in vivo precision errors, 2.1%, 1.5%, and 1.9%, respectively). Hologic QDR 1000/w scanners (Hologic Inc.) were used at baseline but replaced with Hologic QDR 2000 scanners at the following visits. Cross calibration between the centers and the scanners was ensured using the same humanoid phantom and double determination, as described in detail by Abrahamsen et al.(25)

Physical activity, dietary intake, and physical fitness

Physical activity was assessed at baseline, and after 1, 2, and 5 years using a questionnaire assessing the number of hours per week used on walking at work, walking not related to the participants work (during daily activities such as shopping, leisure time, etc.), running (exercise, jogging, etc.), bicycling, dancing, exercising that could not be classified in any of the categories mentioned (e.g., aerobics), and other physical activities not registered in any of the other categories.

The total daily energy intake was assessed using a 7-day food-frequency questionnaire, which was analyzed by a trained dietitian. The analysis of nutrients was done using the Dankost computer program, which holds tables of all food items manufactured and sold in Denmark. The CV for total daily energy intake was 18.9% assessed in a subsample of 15 subjects.

Physical fitness was assessed using cycle ergometry (Ergo-metrics 900 cycle ergometer, ergo-line; Ergoline, Bitz, West Germany(26) at baseline, and after 2 and 5 years. The participants were tested on an electronically braked cycle ergometer with a load that increased automatically by 25 W every 2 minutes. They exercised under this increasing load until exhaustion (maximal exercise). The physical fitness was expressed as the maximal load per kg body weight (e.g., 125 W after 8 minutes in a participant weighing 85 kg would yield a physical fitness of 1.47 W/kg).


Analyses were based on the intention-to-treat principle. A causal analysis was also undertaken in those adhering to their initial allocation to HRT or no HRT. Unless otherwise stated, the causal analysis did not change the results. Mean and SD were used as descriptive statistics. Groups were compared using t-tests for independent samples. Differences between HRT and non-HRT treated over time were analyzed using repeated measures ANOVA. Multiple linear regression analysis was used to study the association between weight changes over the 5 years and changes in physical fitness, energy intake, HRT use, and smoking.

Multiple linear regression analysis was also used to study the association between changes in body weight, LBM, and FM, and changes in BMC. For each visit, models were designed with the bone change as the dependent variable, and the weight change, LBM change, and FM change in the period as predictors. Additional models with LBM, FM, and weight changes in the previous periods were evaluated for the bone changes after 2 and 5 years. To avoid autocorrelation, only weight, FM, and LBM changes for the actual period were allowed in the same model. Adjustment for initial body composition, expressed as fat content in percent of total soft tissue weight, was included in all models. All the models were made separately for the HRT-treated and the untreated subjects and tentatively included a dichotomous variable coding each subject's treatment status as either randomized or self-selected. For the purpose of calculating the effect of HRT adjusted for changes in body composition, nine models—three for each follow-up period —were developed. BMC changes in the lumbar spine, femoral neck, and whole body after each period were predicted in each model from the HRT status and adjusted for changes in LBM, FM, and initial body composition.

The analyses were performed using SPSS 6.1.3 (SPSS, Chicago, IL, USA) or Systat 5.2.1 for the Macintosh (Systat Software, Inc., Richmond, VA, USA).


  1. Top of page
  2. Abstract
  7. References

Among the initial group of 502 women randomized to HRT, 268 (53%) completed the 5 years on unchanged HRT, 43 (9%) changed type of HRT, 137 (27%) terminated treatment, and the remaining 54 did not complete the 5 years (11%). Of the 504 randomized to no HRT, 353 (70%) completed the 5 years without receiving HRT, 96 (19%) at one point or another initiated HRT, and the remaining 55 did not complete the 5 years (11%). Among those on HRT (n = 221) by own choice, 127 (57%) completed 5 years on unchanged HRT, whereas 626 (79%) of the initial 789 women on no HRT completed the 5 years without taking HRT (Table 1).

Table Table 1.. Overview of the Status of the Participants After 5 Years
Thumbnail image of

Table 2 shows baseline characteristics of all participants in the two randomized and two self-selected groups. By chance, women randomized to no HRT were 6 month older than those allocated to HRT. Beyond this, no other differences were observed between the randomized groups. Women choosing HRT were 3.6 kg lighter than those who refused the treatment (p < 0.001). They were also closer to menopause, smoked more frequently, and had a higher frequency of hysterectomy.

Table Table 2.. Baseline Characteristics (Mean and SD) or Actual Number and Percentage
Thumbnail image of

Table 3 details changes in weight, body mass index (BMI), body composition, exercise, working capacity, and energy intake. Small changes over time were observed in all four groups. However, intention-to-treat analysis of changes in body weight showed that subjects randomized to HRT gained less weight during the 5 years than subjects not on HRT. The deficit in weight gain was −0.63 (95% CI, −1.26 to −0.01) kg. The smaller weight gain was almost entirely because of a lesser gain in FM of −0.69 (95% CI, −1.21 to −0.18) kg with no differences in LBM changes over time. The reduced gain of FM in women on HRT was seen both in the trunk and the extremities. However, because of the larger volume of trunk fat, the largest absolute difference occurred in the trunk.

Table Table 3.. Change in Body Weight, Body Composition, Exercise, Work Capacity, and Energy Intake During 5 Years
Thumbnail image of

Despite the differences in weight gain, the absolute weight after 5 years was equal in those randomized to HRT (69.9 ± 12.7 kg) and those randomized to no HRT (69.8 ± 12.2 kg, p = 0.95). In the self-selected group, the weight difference between those opting for HRT (66.4 ± 11.3 kg) and those not opting for HRT remained significant (70.4 ± 12.9 kg, p < 0.01).

As a result of the differences in weight gain, BMI of the HRT-treated increased less than in the non-HRT-treated group. Restricting the analysis to those on either unchanged HRT or no HRT (causal analysis) augmented the differences (mean difference in total body weight: −0.90 [95% CI, −1.70 to −0.10] kg, and mean difference in total body FM: −1.06 [95% CI, −1.71 to −0.41] kg). Women on HRT were a little more physically active than women not on HRT. However, although there were differences in some components of physical activity between the two groups, the total weekly physical activity was not significantly different, and neither was the reported energy intake. Working capacity and energy intake declined with advancing age without significant differences between those on HRT and no HRT. The same patterns were found between the two self-selected groups. The HRT-treated women gained less fat, but both groups decreased their energy intake and their work capacity similarly. The total weekly physical activity decreased in the two self-selected groups and increased in the randomized groups, but the difference was not statistically significant, neither between self-selected and randomized subjects (p = 0.07, not tabulated), nor between treated and untreated subjects (p = 0.30, not tabulated).

An analysis of factors associated with the weight changes showed that a decline in physical fitness was the main determinant of weight gain (Table 3). With each one-unit decrease in physical fitness (W/kg), the weight would increase 2.49 kg over the 5 years, if the woman was in the randomized group, and 1.61 kg, if she was in the self-selected group (Table 4). In the self-selected group, changes in physical activity also influenced body weight. The other determinants (HRT vs. no HRT, smoking, and energy intake) did not influence the weight changes.

Table Table 4.. Factors of Significance to the Change in Weight Over the 5 Years (Multiple Linear Regression)
Thumbnail image of

Bone changes

Models including BMD or BMC gave similar results. The results presented below represent BMC changes. Because overall weight changes had no effect on variations in BMC when adjusted for changes in FM and LBM, weight changes were not included in the following regression models.

BMC changes in the lumbar spine

Table 5 shows the adjusted effects of 1, 2, and 5 years of HRT treatment and the effect of alterations in FM and LBM on lumbar spine BMC (L2-L4), adjusted for the effect of the initial body composition. HRT had a positive influence on BMC amounting to 0.77, 1.21, and 1.72 g after 1, 2, and 5 years, respectively. The effect of the initial relative fat content was of the same size in HRT-treated and -untreated groups, with a clear increasing tendency with time.

Table Table 5.. Changes in the Bone Mineral Content in the Lumbar Spine, Related to Changes in the Lean Body Mass, and in the Fat Mass
Thumbnail image of

In the HRT-treated group, changes in LBM after 1 and 2 years had significant positive effects on lumbar BMC alterations after 1 and 2 years (0.226 g/kg), respectively. However, the largest effect was on the lumbar BMC after 5 years (0.417 g/kg). This effect was of the same order as the effect of LBM changes after 5 years on changes in lumbar spine BMC after 5 years. Changes in FM had no effect on BMC changes in the lumbar spine.

In those not receiving HRT, lumbar BMC changes after 1 and 2 years could not be predicted by changes in LBM or in FM during the same periods. However, FM after 1 and 2 years and LBM after 2 and 5 years yielded significant positive contributions to changes in regional BMC after 5 years.

BMC changes in the femoral neck

Table 6 shows the adjusted effects of 1, 2, and 5 years of HRT treatment, and the effects of alterations in FM and LBM, on femoral neck BMC adjusted for the effect of the initial body composition. HRT had a positive influence on BMC amounting to 42.49, 78.50, and 106.26 mg after 1, 2, and 5 years, respectively. The effects of the initial body composition on femoral neck BMC changes after 1 and 2 years were small and insignificant in all groups, but in the untreated group the effect on bone changes after 5 years was obvious (38.44 mg/%, p < 0.05).

Table Table 6.. Changes in the Bone Mineral Content in the Femoral Neck, Related to Changes in the Lean Body Mass, and the Fat Mass
Thumbnail image of

The changes in femoral neck BMC after 1 and 2 years in the HRT-treated group were best predicted by alterations in FM after 1 and 2 years, whereas BMC changes after 5 years were best predicted from changes in LBM after 2 and 5 years.

In the untreated group, changes in femoral neck BMC after 1 year were related to the corresponding changes in FM but not to changes in LBM. Likewise, after 2 years, BMC changes correlated significantly to changes in FM after 1 and 2 years. After 5 years, there were significant contributions to BMC changes from changes in FM after 1, 2, and 5 years, and from changes in LBM after 2 and 5 years.

Changes in the TBBM

Table 7 shows the adjusted effects of 1, 2, and 5 years of HRT treatment and the effects of alterations in FM and LBM on TBBM adjusted for the effect of the initial body composition. HRT had a positive influence on TBBM amounting to 16.70, 29.28, and 50.65 g after 1, 2, and 5 years, respectively.

Table Table 7.. Changes in the Whole Body Bone Mineral, Related to Changes in the Lean Body Mass and the Fat Mass
Thumbnail image of

The effects of the initial body composition on changes in TBBM after 1 and 2 years were small and insignificant in all groups, but in the untreated group, the effect on bone change after 5 years was clear (1.46 g/%, p = 0.01)

The TBBM changes after 1 year were influenced by changes in FM as well as LBM in both groups. TBBM changes after 2 years depended on changes in FM as well as LBM after 1 and 2 years in both groups. Furthermore, in both groups, TBBM changes after 5 years were influenced by changes in FM as well as LBM after 1, 2, and 5 years.


  1. Top of page
  2. Abstract
  7. References

Changes in weight and body composition

In a large randomized trial, we found a significantly lower weight gain in women on HRT than in women on no HRT. This is in accordance with the nonsignificant trend toward a lower weight gain with estrogen/progestogen preparations reported from the Cochrane collaboration.(14) This meta-analysis reported a difference in weight gain of −0.47 (95% CI, −1.63–0.69 kg) in favor of estrogen/progestogen over placebo in a combined sample of almost the same size as our study sample. The difference in our study amounted to −0.69 kg in the intention to treat analysis and −0.90 kg in the causal analysis, which is somewhat more, but still within the same range as the results from the Cochrane collaboration.(14) One explanation for the discrepancy could be that our study was of longer duration (60 months) than most of the studies in the Cochrane analysis (3–48 months).(14)

Almost the entire difference in weight gain was attributable to differences in accumulation of body fat. The reason for this difference remains unclear. In a subgroup of the patients presented here, no independent influence of HRT on leptin levels and thus appetite regulation could be demonstrated.(27)

Our results of a smaller fat accumulation with HRT are in accordance with the results of Hassager et al.,(10) but in contrast to the findings of Aloia et al.,(28) who reported a greater increase in FM in women randomized to conjugated estrogen (0.625 mg) plus medroxyprogesterone (10 mg) compared with placebo, whereas there was a rapid loss of LBM in both groups. Furthermore, Aloia et al.(28) reported a significant increase in body weight in women on estrogen/progestogen compared with a nonsignificant increase in women on placebo. It should be noted that the results of Aloia et al.(28) concerning a weight gain with HRT are in contrast to the results of the meta-analysis.(14)

The reduced accumulation of body fat took place in the trunk and the extremities, pointing toward a generalized effect on fat tissue of HRT, which was not just restricted to the abdominal fat tissue.

Although the difference in weight gain over 5 years was significant, the final absolute weight was similar in HRT and non-HRT treated. This again underlines that HRT does not lead to an increase in body weight.

Effects of changes in physical fitness and energy intake

As expected, physical fitness and energy intake declined over time. However, HRT did not influence the decline. This is in accordance with the absence of a difference in lean body mass between those on and those not on HRT. The main determinant of the overall weight gain was the age-related decline in physical fitness. After adjustment for physical fitness, the effect of HRT on weight gain was insignificant, but a trend toward a lower weight gain with HRT remained (0.7 kg in intention-to-treat and 0.3 kg in causal analysis). The participants reported having a total weekly physical activity within the same range after 5 years at baseline. This could indicate that, although the physical activity is maintained, it is not able to counteract the age-related decline in working capacity; in other words, had the participants also reduced their weekly physical activity, the weight gain would have been even larger. Although the participants used the same amount of hours per week on activities, they probably did so at a reduced intensity level. A main responsible factor for the weight gain in this middle-aged period thus seems to be a reduced intention or ability to maintain high-intensity work. This may lead to a decline in energy expenditure resulting in an accumulation of body fat, although the energy intake also declines. The observed decline in energy intake with time matches the observation from cross-sectional data.(29) However, the decrease with time may also be due to a change in underreporting of fat intake.(30) These factors may make associations between energy intake and changes in weight uncertain.(31)

Randomized versus observational studies

Women choosing HRT are known to differ from women refusing such treatment.(32–35) As these differences occur in rather crucial points such as educational level, smoking habits, and number of childbirths, the effect of HRT on bone is difficult to evaluate in nonrandomized studies. The present study describes both randomized and self-selected HRT users. We found no difference in the effect of HRT on BMC between these groups.

In the self-selected group, those choosing HRT had a lower initial body weight than those not choosing HRT. However, despite this, the effect of HRT on weight gain was similar in the randomized and nonrandomized study arms. This could indicate that the results from our study can be reliably applied at population level.

Several factors may be involved in the lower initial body weight in the group. First of all, obese women may be less likely to choose HRT because of fear that their weight may increase further, indicating that the myth about a weight-increasing effect of HRT still lives. The women with the lowest body weight are thus less reluctant to opt for HRT.

On the other hand, one could expect that the slimmer women would be more likely to maintain their slimness, but this did not seem to be the case because an increase in body weight was also seen in this group, indicating that other factors besides lifestyle may be involved.

The fact that the women in one study arm could choose HRT themselves could introduce a bias as outlined above. The present study has the advantage to have both a randomized and nonrandomized arm, thus being able to detect differences between populations accepting to participate in a randomized study and those refusing to participate. Any bias from this point can thus be analyzed. In fact no systematic differences could be detected, lending further support to the view that the preventive effect of HRT on weight gain was real and not skewed by the participants own choices.

Because weight loss and low body weight increase the risk for low BMD and fractures, the effect of HRT is likely to be underestimated in uncontrolled studies, if a lower body weight at baseline is not accounted for.

Interactions between the effects of HRT on bone mineral and body composition

In the women in our study not receiving HRT, changes in femoral neck BMC after 1, 2, and 5 years could be predicted from changes in FM. After 5 years, 5-year LBM changes also started to contribute. In the HRT-treated group, on the other hand, the effects of FM changes on BMC alterations were much less pronounced, and the BMC changes after 5 years were only predicted by changes in LBM. This different pattern of influence was also seen in the lumbar spine. Other studies have suggested that bone mass in pre- and perimenopausal women is best predicted from LBM, and that bone mass in untreated postmenopausal women is best predicted from FM.(36–38) These studies have also suggested that, because extragonadal estrogen synthesis in fat tissue is the dominant estrogen source in post- but not in premenopausal women, changes in FM have less influence on bone mass in the latter group. In our study, the effect of FM on femoral neck and lumbar spine BMC was markedly reduced in women receiving HRT compared with those who did not. This observation supports the above-mentioned theory. However, we still need a study, which will clearly show that changes in plasma levels estrogen are better predictors of alterations in BMC than FM. Also, it cannot be excluded that other factors such as mechanical loading and plasma leptin levels,(39–41) which both depend on body mass and composition, may modulate the effect of FM and LMT on bone tissue.(39–41) However, so far, the observed relations between leptin levels and bone in adults seem more to be a result of the corresponding fat changes than being the consequences of a direct effect of leptin on the skeleton.(39, 41)

In our study, changes in LBM affected bone mass in the same direction and magnitude as changes in FM, but the effect was not reduced by HRT. On the contrary, we found that HRT increased the effect of LBM on BMC (from 0.123 to 0.417 g/kg; Table 5). In this study, HRT did not increase the average LBM, neither in the randomized nor in the self-selected group. However, other studies(10) have suggested individual LBM changes to be markers of the biological tissue impact of HRT. A larger LBM increase during HRT would then indicate that this particular person had an increased susceptibility for HRT, for instance through a greater bioavailability. Such a mechanism would explain why LBM and HRT seem to have additive effects on bone mass.

As previously described,(9) the effect of HRT on BMC in our study could not be explained solely by changes in body composition. Five years of HRT resulted in a positive effect on lumbar BMD equivalent to the calculated effect of a 14-kg increase in LBM in an untreated woman. The corresponding HRT effects on the femoral neck and the TBBM were equivalent to 4.5 and 2.5 kg of LBM, respectively. However, the average changes in LBM over the 5-year period in the four groups were 10-fold smaller than this, that is, less than 300 g. Thus, a direct effect of HRT still plays the key role for the bone changes observed after HRT in early postmenopausal women.

In conclusion, HRT may to some degree prevent the accumulation of fat in early postmenopausal women related to a decreased physical fitness. In the absence of HRT, this fat accumulation to some extent protects against bone loss. Changes in LBM were unrelated to HRT, but loss of LBM had a negative influence on BMC.


  1. Top of page
  2. Abstract
  7. References
  • 1
    Køster A 1990 Hormone replacement therapy: Use patterns in 51-year-old Danish women. Maturitas 12:345356.
  • 2
    Munk-Jensen N, Nielsen SP, Obel EB, Eriksen PB 1988 Reversal of postmenopausal vertebral bone loss by oestrogen and progestogen: A double blind controlled study. BMJ 296:11501152.
  • 3
    Quigley TEM, Purvis LM, Burnier AM, Brooks P 1987 Estrogen therapy arrests bone loss in elderly women. Am J Obstet Gynecol 156:15161523.
  • 4
    Lindsay R 1980 Prevention of spinal osteoporosis in oophorectomised women. Lancet 2:11511154.
  • 5
    Lindsay R 1993 Consensus development conference on osteoporosis; hormone replacement and treatment for osteoporosis. Am J Med 95:1S78S.
  • 6
    Paganini-Hill A, Ross RK, Gerkins VR, Henderson BE, Arthur M, Mack TM 1981 Menopausal estrogen therapy and hip fractures. Ann Intern Med 95:2831.
  • 7
    Ettinger B, Genant HK, Cann CE 1985 Long-term estrogen replacement therapy prevents bone Loss and fractures. Ann Intern Med 102:319324.
  • 8
    Lufkin EG, Wahner HW, O'Fallon WM, Hodgson SF, Kotowicz MA, Lane AW, Judd HL, Caplan RH, Riggs BL 1992 Treatment of postmenopausal osteoporosis with transdermal estrogen. Ann Intern Med 117:19.
  • 9
    Mosekilde L, Beck-Nielsen H, Sørensen OH, Nuelsen SP, Charles P, Vestergaard P, Hermann AP, Gran J, Hansen TB, Abrahamsen B, Ebbesen EN, Stilgren L, Jensen LB, Brot C, Hansen B, Tofteng CL, Eiken P, Kolthoff W 2000 Hormonal replacement therapy reduces forearm fracture incidence in recent postmenopausal women-results of the Danish Osteoporosis Prevention Study. Maturitas 36:181193.
  • 10
    Hassager C, Christiansen C 1989 Estrogen/gestagen therapy changes soft tissue body composition in postmenopausal women. Metabolism 38:662665.
  • 11
    Panotopoulos G, Raison J, Ruiz JC, Guy Grand B, Basdevant A 1997 Weight gain at the time of menopause. Hum Reprod 12(Suppl 1):126133.
  • 12
    Nachtigall LE 1990 Enhancing patient compliance with hormone replacement therapy at menopause. Obstet Gynecol 75(Suppl 4):77S80S.
  • 13
    Ryan PJ, Harrison R, Blake GM, Fogelman I 1992 Compliance with hormone replacement therapy (HRT) after screening for post menopausal osteoporosis. Br J Obstet Gynacol 99:325328.
  • 14
    Norman RJ, Flight IH, Rees MC 2000 Oestrogen and progestogen hormone replacement therapy for peri-menopausal and post-menopausal women: Weight and body fat distribution. Cochrane Database Syst Rev 2:CD001018CD001018.
  • 15
    van Seumeren I 2000 Weight gain and hormone replacement therapy: Are women's fears justified? Maturitas 34(Suppl 1):S3S8.
  • 16
    Chmouliovsky L, Habicht F, James RW, Lehmann T, Campana A, Golay A 1999 Beneficial effect of hormone replacement therapy on weight loss in obese menopausal women. Maturitas 32:147153.
  • 17
    Crawford SL, Casey VA, Avis NE, McKinlay SM 2000 A longitudinal study of weight and the menopause transition: Results from the Massachusetts Women's Health Study. Menopause 7:96104.
  • 18
    Reubinoff BE, Wurtman J, Rojansky N, Adler D, Steen P, Schenker JG, Brzezinski A 1995 Effects of hormone replacement therapy on weight, body composition, fat distribution, and food intake in early postmenopausal women: A prospective study. Fertil Steril 64:963968.
  • 19
    Espeland MA, Stefanick ML, Kritz-Silverstein D, Fineberg SE, Waclawiw MA, James MK, Greendale GA 1997 Effect of postmenopausal hormone therapy on body weight and waist and hip girths. Postmenopausal Estrogen-Progestin Interventions Study Investigators. J Clin Endocrinol Metab 82:15491556.
  • 20
    Compston J, Laskey M, Croucher P, Coxon A, Kreitzman S 1992 Effect of diet-induced weight loss on total body bone mass. Clin Sci (Lond) 82:429432.
  • 21
    Jensen LB, Quaade F, Sørensen OH 1994 Bone loss accompanying voluntary weight loss in obese humans. J Bone Miner Res 9:459463.
  • 22
    Ricci TA, Chowdhury HA, Heymsfield SB, Stahl T, Pierson RN Jr, Shapses SA 1998 Calcium supplementation suppresses bone turnover during weight reduction in postmenopausal women. J Bone Miner Res 13:10451050.
  • 23
    Mosekilde L, Hermann AP, Beck-Nielsen H, Charles P, Nielsen SP, Sørensen OH 1999 The Danish Osteoporosis Prevention Study (DOPS): Project design and inclusion of 2016 normal perimenopausal women. Maturitas 31:207219.
  • 24
    Vestergaard P, Hermann AP, Gram J, Jensen LB 1997 Improving compliance with hormonal replacement therapy in primary osteoporosis prevention. Maturitas 28:137145.
  • 25
    Abrahamsen B, Gram J, Hansen TB, Beck-Nielsen H 1995 Cross-calibration of QDR-2000 and QDR-1000 dual energy X-ray densitometers for bone mineral and soft tissue measurements. Bone 16:385391.
  • 26
    Petersen JR, Drabaek H, Mehlsen J, Haedersdal C, Moller ML 1993 Systolic blood pressure at rest and during submaximal and maximal exercise on ergometer bicycle. The value of automatic and auscultatory determination. Ugeskr Laeger 155:631635.
  • 27
    Kristensen K, Pedersen SB, Vestergaard P, Mosekilde L, Richelsen B 1999 Hormone replacement therapy affects body composition and leptin differently in obese and non-obese postmenopausal women. J Endocrinol 163:5562.
  • 28
    Aloia JF, Vaswani A, Russo L, Sheehan M, Flaster E 1995 The influence of menopause and hormonal replacement therapy on body cell mass and body fat mass. Am J Obstet Gynecol 172:896900.
  • 29
    Andersen NL, Fagt S, Groth MV, Hartkopp HB, Møller A, Ovesen L, Warming DL 1995 Danskernes Kostvaner. Levnedsmiddelstyrelsen, Copenhagen, Denmark, pp. 1298.
  • 30
    Heitmann BL, Lissner L, Osler M 2000 Do we eat less fat, or just report so? Int J Obes Relat Metab Disord 24:435442.
  • 31
    Jørgensen LM, Sørensen TI, Schroll M, Larsen S 1995 Influence of dietary factors on weight change assessed by multivariate graphical models. Int J Obes Relat Metab Disord 19:909912.
  • 32
    Derby CA, Hume AL, Barbour MM, McPhillips J, Lasater TM, Carleton RA 1993 Correlates of post-menopausal estrogen use and trends through the 1980s in two south-eastern New England communities. Am J Epidemiol 137:11251135.
  • 33
    Oddens BJ, Boulet MJ, Lehert P, Visser AP 1992 Has the climacteric been medicalized? A study on the use of medication for climacteric complaints in four countries. Maturitas 15:171181.
  • 34
    MacLennan AH, MacLennan A, Wilson D 1993 The prevalence of oestrogen replacement therapy in South Australia. Maturitas 16:175183.
  • 35
    Jensen LB, Hilden J 1996 Sociological and behavioural characteristics of perimenopausal women with an express attitude to hormone substitution therapy. Maturitas 23:7383.
  • 36
    Reid IR, Legge M, Stapleton JP, Evans MC, Grey AB 1995 Regular exercise dissociates fat mass and bone density in premenopausal women. J Clin Endocrinol Metab 80:17641768.
  • 37
    Salamone LM, Glynn N, Black D, Epstein RS, Palermo L, Meilahn E, Kuller LH, Cauley JA 1995 Body composition and bone mineral density in premenopausal and early perimenopausal women. J Bone Miner Res 10:17621768.
  • 38
    Brot C, Jensen LB, Sørensen OH 1997 Bone mass and risk factors for bone loss in perimenopausal Danish women. J Intern Med 242:505511.
  • 39
    Goulding A, Taylor RW 1998 Plasma leptin values in relation to bone mass and density and to dynamic biochemical markers of bone resorption and formation in postmenopausal women. Calcif Tissue Int 63:456458.
  • 40
    Klein KO, Larmore KA, de Lancey E, Brown JM, Considene RV, Hassink SG 1998 Effect of obesity on estradiol level, and its relationship to leptin, bone maturation, and bone mineral density in children. J Clin Endocrinol Metab 83:34693475.
  • 41
    Odabasi E, Ozata M, Turan M, Bingol N, Yonem A, Cakir B, Kutlu M, Ozdemir K 2000 Plasma leptin concentrations in postmenopausal women with osteoporosis. Eur J Endocrinol 142:170173.