Journal Logo


Influence of Estradiol Status on Physical Activity in Premenopausal Women


Author Information
Medicine & Science in Sports & Exercise: August 2018 - Volume 50 - Issue 8 - p 1704-1709
doi: 10.1249/MSS.0000000000001598
  • Free


Preclinical studies have demonstrated that the loss of estradiol (E2) or disruption of E2 signaling causes excess fat gain in female rodents, particularly in abdominal regions (1–4). In ovariectomized animals, this is prevented by E2 treatment (5). Similarly, suppression of gonadal function by gonadotropin-releasing hormone agonist (GnRHAG) therapy in humans causes excess fat gain (6–10) with abdominal fat accrual (11). In both women (11) and men (8), limited evidence indicates that such changes are specifically related to E2 deficiency. One mechanism that may contribute to excess fat gain in response to E2 suppression is a reduction in energy expenditure (EE). In animals, the gain in fat mass resulting from loss of E2 or E2 signaling has been associated with decreases in both resting (REE) and total EE (TEE) (12–16). In premenopausal women, we have previously shown that acute (6 d) and chronic (5 months) suppression of ovarian hormones caused decreases in REE (17,18) and TEE (18), and the decrease in REE was prevented by E2 treatment (18).

The decrease in TEE with loss of E2 or disruption in E2 signaling is due, in part, to a decline in spontaneous physical activity (SPA) and the associated EE. Both ovariectomy (3,19) and the disruption of E2 receptor signaling (4) in rodents cause a dramatic decline in SPA that is fully rescued by E2 treatment (20). However, the effects of E2 on physical activity (PA) in humans have not been extensively explored. In National Health and Nutrition Examination Survey III, the prevalence of women reporting being physical inactive was lower in women who ever used hormone therapy (HT: 28.5%) when compared with those who had never used HT (40.0%) (21). We are aware of only one study that obtained objective measures of PA across the menopause transition; in that study, PA at the time of menopause decreased by approximately 50% compared with 3–4 yr before menopause (22). However, whether the decline in PA was related to the loss of E2, or instead biological aging or other factors associated with the menopausal transition could not be determined.

The goal of this study was to determine the effects of 5 months of ovarian hormone suppression in the absence or presence of E2 add-back on objectively measured PA. Because PA was an exploratory outcome in the parent trial, which evaluated changes in adiposity, REE, and TEE (11,18), the study was not powered to detect changes in PA. Therefore, these data represent a proof-of-concept investigation to evaluate whether ovarian hormone suppression is associated with a decrease in free-living PA, as occurs in rodents.


Institutional approval

This study was approved by the Colorado Multiple Institutional Review Board and the Scientific Advisory and Review Committee at the University of Colorado Anschutz Medical Campus. The study was registered on (NCT00687739) on May 28, 2008.

Participants and screening procedures

Participants were healthy premenopausal women. Volunteers provided written informed consent to participate, with the knowledge that the risks of the study included menopause-like effects (e.g., weight gain, bone loss, and menopausal symptoms). Volunteers underwent screening procedures, as previously described (11,18). Primary inclusion criteria were age (25–49 yr) and normal menstrual cycle function (no missed cycles in previous year, cycle length of 28 ± 5 d, and confirmation of ovulatory status (ClearPlan Easy; Unipath Diagnostics, Waltham, MA)). Primary exclusion criteria were pregnancy or lactation; use of hormonal contraception, oral glucocorticoids, or diabetes medications; smoking; or body mass index (BMI) of >39 kg·m−2.

Experimental design and study procedures

The parent trial was a randomized, double-blinded, placebo-controlled trial to determine the effects of E2 deficiency on body composition and components of EE (11,18). All participants underwent suppression of ovarian sex hormones with GnRHAG therapy (leuprolide acetate 3.75 mg, Lupron; TAP Pharmaceutical Products, Inc, Lake Forest, IL), delivered as monthly intramuscular injections. A single injection of leuprolide acetate produces an initial stimulation (for 1–3 wk) followed by a prolonged suppression of pituitary gonadotropins, and repeated monthly dosing suppresses ovarian hormone secretion (23). Absence of pregnancy (urine test) was confirmed before each dosing. Participants were randomized to receive concurrent transdermal E2 0.075 mg·d−1 (Bayer HealthCare Pharmaceuticals, Berkeley, CA) or placebo patches (GnRHAG + E2, n = 35; GnRHAG + PL, n = 35). During these monthly visits, participants met with the research nurse practitioner. Participants were queried about changes in use of medications or health (e.g., doctor visits and hospitalizations), as well as any study-related problems/concerns over the past 4 wk. Compliance to the transdermal patches was confirmed verbally each month during the monthly examination. The E2 regimen was expected to maintain serum E2 concentrations in the mid-to-late follicular phase range (100–150 pg·mL−1).

As previously described (11,18), some participants in each drug group in the parent study were also randomized to a supervised progressive resistance exercise training program (N = 12 in each group) to attenuate the expected decrease in fat-free mass with GnRHAG therapy. Not all those randomized to exercise were willing to participate in the exercise program and were therefore treated as nonexercisers for the analyses herein. The progressive resistance exercise intervention consisted of 4 d·wk−1 for 18 wk beginning in week 1 of sex hormone suppression (11).


PA was assessed using a hip-worn accelerometer (Actical; Mini Mitter Co., Inc., Bend, OR). Sampling was performed at 32 Hz and data were stored in 1-min epochs. The monitor measures accelerations in multiple planes but is most sensitive to vertical movements of the torso when worn on the hip. The resulting raw output is activity counts, which represent both the summed quantity and magnitude of accelerations during each epoch. PA was assessed for 1 wk before the intervention and for 1 wk of each month of the 5-month intervention. Participants were instructed to wear the Actical at all times for 7 consecutive days, except when engaged in water activities (e.g., bathing, showering, or swimming) or sleeping. To be included in the final analysis, 4 d of valid data was required for each time point. A day was considered valid if there was at least 10 h of wear time.

Estimates of time spent in sedentary, light, and moderate-to-vigorous activity (MVPA) were derived using thresholds for number of accelerometer activity counts accumulated in a given epoch for the Actical in middle-age adults (24). To assess whether the levels of MVPA met the 2008 Guidelines for Physical Activity in Americans (25), we examined the number of bouts lasting at least 10 min in duration with >80% of the bout categorized as MVPA (guideline MVPA) (25). Guideline MVPA was calculated as both minutes per day (total minutes per day spent in MVPA bouts) and the number of bouts per day (the number of MPVA bouts per day). For participants in the nonexercise groups, we also determined the amount of time spent in SPA, defined as all minutes spent in light PA or MVPA performed in bouts lasting less than 10 min.

Sex hormones

Blood samples for sex hormones were collected during baseline testing and during week 20 of the intervention. A single sample (~5 mL) was obtained in the morning (~8 am), after an overnight fast (at least 10 h), and at least 48 h after the last bout of exercise. Baseline samples were obtained immediately before the first injection. Collection samples were stored at −80°C until analysis. E2, progesterone, and total testosterone were analyzed using chemiluminescence immunoassay with intraday and interday coefficients of variation of 4.3% and 8.2% for E2, 4.4% and 7.9% for progesterone, 3.6% and 5.7% for sex hormone–binding globulin (SHBG), and 2.1% and 5.1% for total testosterone (Access 2 Immunoassay System Analyzer; Beckman Coulter, Fullerton, CA).

Statistical analysis

Baseline characteristics are presented as mean (SD) by group. Changes in these characteristics in response to the intervention are presented as mean (95% confidence interval (CI)). The primary analysis evaluated differences in time spent in sedentary behavior and in light PA and MVPA between the GnRHAG + PL and GnRHAG + E2 groups using a repeated-measures maximum likelihood and cell means model regressing activity levels on group and time, controlling for baseline, using all available data. This approach is similar to a repeated-measures ANCOVA, but has the advantage of not requiring casewise deletion of cases missing information at one or more time points. To minimize the number of statistical tests, we estimated the overall group difference, the overall time effect, and the difference between groups at the month 4 time point. This time point was chosen because it was the last full month of the exercise intervention; participants in the exercise intervention stopped exercising 2 wk before the follow-up assessments at month 5 to allow body weight and other metabolic outcomes to stabilize before the postintervention assessments.

Because participation in the supervised exercise program could have influenced PA performed outside of the programmed exercise time, we also evaluated the effects of the intervention on the subset of women in each group who did not participate in the supervised exercise program. The maximum number of participants with valid PA data was 16 per group at visit 2; the minimum was 11 GnRHAG + E2 and 14 GnRHAG + PL at visit 6. All analyses were performed using SAS 9.3 (SAS Institute, Cary, NC). Data are reported as mean (SD) or mean (95% CI), unless otherwise specified.


Descriptive characteristics of participants in the GnRHAG + E2 (N = 31) and GNRHAG + PL (N = 30) groups who had any PA data (61 of 70 participants in the parent trial) are shown in Table 1. At baseline, the groups were similar in age, weight, BMI, and body composition. As in the larger cohort (11), fat-free mass decreased over 5 months in the GnRHAG + PL but not the GnRHAG + E2 group; no significant changes in weight, BMI, or fat mass were observed in either group over this time frame (11). Changes in sex steroid hormones were similar to those observed in the larger cohort. E2 decreased in GnRHAG + PL (from 85.6 ± 41.2 to 21.8 ± 8.5 pg·mL−1; P < 0.001) and increased in GnRHAG + E2 (61.7 ± 46.5 to 99.2 ± 93.2 pg·mL−1; P = 0.09). Although we did not reach target, response to E2 add-back was highly variable. Nonetheless, the between-group changes were significant. Progesterone and testosterone decreased in both groups (all P < 0.03), whereas SHBG decreased only in GnRHAG + PL (P < 0.01).

Baseline characteristics (mean (SD)) of participants randomized to GnRHAG plus add-back of placebo (GnRHAG + PL, N = 31) or E2 (GnRHAG + E2, N = 30) and changes after 5 months of treatment (mean (95% CI)).

Minutes spent in sedentary behavior and in different levels of PA for all women is shown in Table 2. Sedentary time decreased (P = 0.02), but MVPA (P = 0.03; Fig. 1) and SPA (P = 0.01; Fig. 2) increased in both groups during the intervention. At month 4, there were no differences between groups in sedentary time or SPA, but MVPA tended toward being higher in the GnRHAG + E2 group. There were no significant effects of intervention or time in light PA, guideline MVPA, or the number of guideline MVPA bouts.

For all subjects: time (min·d−1) recorded in sedentary behavior, light PA, and MVPA; guideline MVPA (total minutes spent in MVPA during bouts ≥10 min); number of guideline bouts; and SPA (total minutes spent in light or MVPA performed in bouts lasting <10 min).
Total MVPA in participants randomized to GnRHAG plus add-back of placebo (GnRHAG + PL, N = 31) or E2 (GnRHAG + E2, N = 30). *P < 0.05 at month 4 time point. BL, baseline.
SPA in participants randomized to GnRHAG plus add-back of placebo (GnRHAG + PL, N = 31) or E2 (GnRHAG + E2, N = 30). SPA tended toward being higher in GnRHAG + E2 (P = 0.08), but there were no differences between groups at month 4 time point. BL, baseline.

PA data for the 28 women who did not participate in the supervised exercise program are presented in Table 3. Actical data on 16 nonexercising participants in each group were obtained. No significant time effects were observed in any of the PA outcomes. Total time in MVPA (Fig. 3) tended toward being higher in GnRHAG + E2 (P = 0.07) compared with GnRHAG + PL, but there were no group differences in any other measure. At month 4, MVPA (P = 0.01) and SPA (P = 0.01) were significantly higher in GnRHAG + E2.

For nonexercisers: time (min·d−1) recorded in sedentary behavior, light PA, and MVPA; guideline MVPA (total minutes spent in MVPA during bouts ≥10 min); number guideline bouts; and SPA (total minutes spent in light or MVPA performed in bouts lasting <10 min).
Total MVPA in nonexercising participants randomized to GnRHAG plus add-back of placebo (GnRHAG + PL, N = 16) or E2 (GnRHAG + E2, N = 16). *P < 0.05 at month 4 time point. BL, baseline.


Studies in animals (1,2,4,16) and humans (6–11) provide clear evidence that loss of estrogen leads to weight gain and accumulation of central fat (11). Although the mechanisms contributing to these changes in body fat mass and distribution are not entirely understood, studies of rodents have demonstrated that E2 signaling may play an important role in regulating SPA (3,16,19,20). However, isolating the effects of E2 on PA in humans is very challenging. In this preliminary investigation, we demonstrated that although women in the GnRHAG + PL and GnRHAG + E2 treatment arms had similar levels of PA at baseline and at 1 month, women in the GnRHAG + E2 group had higher levels MVPA during months 2–5 of treatment (Figs. 1 and 2). This divergence coincides with the time frame at which GnRHAG treatment would be expected to be exerting maximal effects to suppress ovarian function. A single injection of leuprolide acetate produces an initial stimulation of GnRH (for up to 3 wk) followed by a prolonged suppression of anterior pituitary gonadotropins and gonadal sex steroids. Repeated monthly dosing maintains ovarian hormone suppression. We observed similar trends in MVPA when women who were not randomized to the supervised exercise program were analyzed separately. Because this study was not powered on PA outcomes, positive or negative findings should be interpreted cautiously. Nonetheless, these data provide the first proof-of-concept evidence that E2 status may contribute to the regulation of PA in women as it does in rodents.

Although the mechanism is not completely understood, some studies suggest that E2 influences PA via ERα-mediated alterations in dopaminergic activity and function (26). Although some evidence exists that PA declines across the menopause (23) and that PA is higher in women using HT (21), convincing evidence in humans is lacking. The latter study was limited by the use of self-report instruments to measure PA as well as the variety of HT regimens used by participants, which prevented the specific effects of E2 to be determined. The best evidence to date supporting a role of sex steroids in regulating PA in humans comes from the study of Lovejoy et al. (22). Premenopausal women (N = 103) who were approaching menopause were studied annually for 4 yr. Menopausal status was confirmed by cessation of menstruation and follicle-stimulating hormone concentrations of ≥30 mIU·mL−1. Accelerometer counts dropped by ~50% over the 4 yr before postmenopausal status was confirmed. However, it was still not possible to isolate the specific effects of E2 on PA. In this study, we show that in women who are undergoing suppression of ovarian function, replacing E2 is associated with maintenance of higher levels of PA compared with women treated with placebo.

Although the PA outcomes in this study were designed to be exploratory, the randomized controlled study-design, double-blinding, robust pharmacological approach to control the sex steroid milieu, and the objective monitoring of PA are strengths that lend support to our conclusions. The conclusions should also be interpreted in light of several limitations. First, monitoring PA using a monitor worn on the hip is not well suited for distinguishing sedentary behavior. Devices worn on the thigh are more sensitive to posture and perform better in classifying sedentary behavior (27). However, these devices were not widely available when this study was conceived. Thus, it is possible that there were differences between treatment arms in the allocation between sedentary behavior and light physical activities that were not detected. Second, GnRHAG therapy does not completely reflect the changes that occur during the natural menopause transition. GnRHAG results in a more abrupt suppression of sex hormones than menopause, and gonadotropins are suppressed rather than elevated. It should be noted that there was a wide range in response to the E2 add-back therapy. We targeted mid-to-late follicular phase concentrations (100–150 pg·mL−1), and mean E2 concentrations in the GnRHAG + E2 group at month 5 was 99 ± 93 pg·dL−1. However, mean E2 concentrations at month 5 in the GnRHAG + PL group was 21 ± 9 pg·mL−1. Thus, although the mean E2 concentrations were slightly below the targeted concentrations, the mean E2 concentrations in the two groups at 5 months differed substantially. Finally, our study was limited to women, and whether E2 would have similar effects on PA in men is not known. Such investigations are warranted given the findings that E2 also contributes to body fat regulation in men (8). In that study, suppressing gonadal function in men combined with aromatase inhibition to suppress estrogen synthesis caused an increase in visceral fat, whereas suppression of gonadal function without aromatase inhibition had no effect on visceral fat. The mechanisms contributing to this gain in fat in men with E2 suppression have not been elucidated.


The results of this proof-of-concept study provide support for the hypothesis that PA levels are maintained at a higher level in women undergoing pharmacological suppression of ovarian function with E2 add-back when compared with women treated with placebo add-back. This provides the first preliminary evidence for the regulation of PA by E2 in women. Given the exploratory nature of this study, confirmatory investigations will be necessary.

We are grateful to the nursing, bionutrition, core laboratory, information systems, and administrative staffs of the Clinical and Translational Research Center and Energy Balance Core of the Nutrition and Obesity Research Center for their support of the study. We also acknowledge the members of our research group who carried out the day-to-day activities for the project. Finally, we thank the women who volunteered to participate in the study for their time and efforts.

This work was supported by National Institutes of Health grants P50 HD073063, R01 AG018198, P30 DK048520, UL1 TR001082, and K01 DK109053. Drs. Melanson, Kohrt, and Schwartz are also supported by resources from the Geriatric Research, Education, and Clinical Center at the Denver VA Medical Center.

The authors have no conflicts to declare. The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation, and statement that results of the present study do not constitute endorsement by the American College of Sports Medicine.

The contents do not represent the views of the US Department of Veterans Affairs or the US Government.


1. Heine PA, Taylor JA, Iwamoto GA, Lubahn DB, Cooke PS. Increased adipose tissue in male and female estrogen receptor-alpha knockout mice. Proc Natl Acad Sci U S A. 2000;97(23):12729–34.
2. Wade GN, Gray JM, Bartness TJ. Gonadal influences on adiposity. Int J Obes. 1985;9(1 Suppl):83–92.
3. Witte MM, Resuehr D, Chandler AR, Mehle AK, Overton JM. Female mice and rats exhibit species-specific metabolic and behavioral responses to ovariectomy. Gen Comp Endocrinol. 2010;166(3):520–8.
4. Xu Y, Nedungadi TP, Zhu L, et al. Distinct hypothalamic neurons mediate estrogenic effects on energy homeostasis and reproduction. Cell Metab. 2011;14(4):453–65.
5. Wohlers LM, Spangenburg EE. 17Beta-estradiol supplementation attenuates ovariectomy-induced increases in Atgl signaling and reduced perilipin expression in visceral adipose tissue. J Cell Biochem. 2010;110(2):420–7.
6. Douchi T, Kuwahata R, Yamasaki H, et al. Inverse relationship between the changes in trunk lean and fat mass during gonadotropin-releasing hormone agonist therapy. Maturitas. 2002;42(1):31–5.
7. Douchi T, Kuwahata T, Yoshimitsu N, Iwamoto I, Yamasaki H, Nagata Y. Changes in serum leptin levels during Gnrh agonist therapy. Endocr J. 2003;50(3):355–9.
8. Finkelstein JS, Yu EW, Burnett-Bowie SA. Gonadal steroids and body composition, strength, and sexual function in men. N Engl J Med. 2013;369(25):2457.
9. Revilla R, Revilla M, Villa LF, Cortes J, Arribas I, Rico H. Changes in body composition in women treated with gonadotropin-releasing hormone agonists. Maturitas. 1998;31(1):63–8.
10. Yamasaki H, Douchi T, Yamamoto S, Oki T, Kuwahata R, Nagata Y. Body fat distribution and body composition during Gnrh agonist therapy. Obstet Gynecol. 2001;97(3):338–42.
11. Shea KL, Gavin KM, Melanson EL, et al. Body composition and bone mineral density after ovarian hormone suppression with or without estradiol treatment. Menopause. 2015;22(10):1045–52.
12. Camporez JP, Jornayvaz FR, Lee HY, et al. Cellular mechanism by which estradiol protects female ovariectomized mice from high-fat diet-induced hepatic and muscle insulin resistance. Endocrinology. 2013;154(3):1021–8.
13. da Silva RP, Zampieri TT, Pedroso JA, et al. Leptin resistance is not the primary cause of weight gain associated with reduced sex hormone levels in female mice. Endocrinology. 2014;155(11):4226–36.
14. Litwak SA, Wilson JL, Chen W, et al. Estradiol prevents fat accumulation and overcomes leptin resistance in female high-fat diet mice. Endocrinology. 2014;155(11):4447–60.
15. Mamounis KJ, Yang JA, Yasrebi A, Roepke TA. Estrogen response element-independent signaling partially restores post-ovariectomy body weight gain but is not sufficient for 17β-estradiol’s control of energy homeostasis. Steroids. 2014;81:88–98.
16. Mitsushima D, Takase K, Funabashi T, Kimura F. Gonadal steroids maintain 24 h acetylcholine release in the hippocampus: organizational and activational effects in behaving rats. J Neurosci. 2009;29(12):3808–15.
17. Day DS, Gozansky WS, Van Pelt RE, Schwartz RS, Kohrt WM. Sex hormone suppression reduces resting energy expenditure and {beta}-adrenergic support of resting energy expenditure. J Clin Endocrinol Metab. 2005;90(6):3312–7.
18. Melanson EL, Gavin KM, Shea KL, et al. Regulation of energy expenditure by estradiol in premenopausal women. J Appl Physiol (1985). 2015;119(9):975–81.
19. Rogers NH, Perfield JW 2nd, Strissel KJ, Obin MS, Greenberg AS. Reduced energy expenditure and increased inflammation are early events in the development of ovariectomy-induced obesity. Endocrinology. 2009;150(5):2161–8.
20. Gorzek JF, Hendrickson KC, Forstner JP, Rixen JL, Moran AL, Lowe DA. Estradiol and tamoxifen reverse ovariectomy-induced physical inactivity in mice. Med Sci Sports Exerc. 2007;39(2):248–56.
21. Andersen RE, Crespo CJ, Franckowiak SC, Walston JD. Leisure-time activity among older US women in relation to hormone-replacement-therapy initiation. J Aging Phys Act. 2003;11(1):82–9.
22. Lovejoy JC, Champagne CM, de Jonge L, Xie H, Smith SR. Increased visceral fat and decreased energy expenditure during the menopausal transition. Int J Obes (Lond). 2008;32(6):949–58.
23. Belchetz PE, Plant TM, Nakai Y, Keogh EJ, Knobil E. Hypophysial responses to continuous and intermittent delivery of hypopthalamic gonadotropin-releasing hormone. Science. 1978;202(4368):631–3.
24. Klippel NJ, Heil DP. Validation of energy expenditure prediction algorithms in adults using the Actical activity monitor. Med Sci Sports Exerc. 2003;35:S284.
25. United States Department of Health and Human Services. 2008 Physical Activity Guidelines for Americans: Be Active, Healthy, and Happy! Washington (DC): U.S. Dept. of Health and Human Services; 2008. pp. ix, 61.
26. Lightfoot JT. Sex hormones’ regulation of rodent physical activity: a review. Int J Biol Sci. 2008;4(3):126–32.
27. Steeves JA, Bowles HR, McClain JJ, et al. Ability of thigh-worn Actigraph and Activpal monitors to classify posture and motion. Med Sci Sports Exerc. 2015;47(5):952–9.


Copyright © 2018 by the American College of Sports Medicine