Controversy exists regarding the role of estrogen replacement on skeletal muscle in postmenopausal women. In 1993, Phillips and colleagues (28) stimulated interest in this area when they reported a decline in the specific force of the adductor pollicis muscle in women around the time of menopause that was not evident in those using hormone replacement. However, subsequent studies have provided conflicting results with a number indicating that estrogen either preserves or enhances muscle performance (17,20,33,34) and others reporting no association (4,6,11,31,35).
Similarly, muscle mass has been reported to be enhanced with hormone replacement in postmenopausal women, combined with or without exercise (33), whereas others have reported no effect (6,11). The mechanisms that underlie the purported beneficial effect of estrogen in skeletal muscle are also unclear. Estrogen may have a direct effect on muscle via estrogen receptors (24) and an indirect effect via the somatotropic axis (12). Estrogen also regulates carbohydrate and lipid metabolism (9), sparing muscle glycogen and stimulating lipid oxidation. As a result, skeletal muscle composition may be adversely affected in postmenopausal women not on estrogen replacement, which would compromise force development and, in turn, physical function.
Given the recent adverse cardiovascular findings of estrogen-alone and estrogen plus progestin treatment from the Womens’ Health Initiative (3,30), it seems appropriate to fully investigate what additional benefits to those already known, such as relief of menopausal symptoms and reduction in fracture risk (5), may be derived from estrogen replacement. The purpose of this analysis was to determine whether muscle composition, as determined by computer tomography (CT) scanning of the midthigh, is superior in current users of estrogen replacement therapy (ERT) compared with nonusers, and whether users also had enhanced muscle strength and physical function. Parameters of muscle composition assessed were muscle and fat cross-sectional area (CSA) and muscle attenuation, a noninvasive measure of muscle density in which lower values reflect a higher lipid content (14). We hypothesized that estrogen users would have superior muscle composition (i.e., greater muscle CSA, higher muscle attenuation, lower fat CSA) due to the hormonal and metabolic effects of estrogen, and as a result have higher levels of muscle strength and physical function.
Subjects were 840 white women aged 70–79 yr, participating in the baseline examination of the Health, Aging, and Body Composition (Health ABC) Study, a multicenter prospective cohort study investigating changes in body composition and health conditions with incident disability in well-functioning, community-dwelling, ambulatory elders. Participants were recruited from two field centers located at the University of Pittsburgh and the University of Tennessee, Memphis. The full Health ABC cohort consists of 3075 individuals and included 855 white women and 729 black women. Black women were not included in the analysis due to the small number (N = 84) currently using ERT. In addition, 15 white women were not included because they were taking androgens (N = 12; 5 androgens alone, 7 androgens and ERT) or unopposed progestogen (N = 3). Recruitment was undertaken from a random sample of white Medicare beneficiaries in selected zip codes surrounding the field centers. To be eligible for Health ABC, participants had to report no difficulty walking a quarter of a mile, climbing 10 steps without resting, or undertaking activities of daily living; be free of life-threatening cancers with no active treatment within the past 3 yr; and planning to remain within the study area for at least 3 yr. Estrogen usage (and progestogen usage) was derived from drug data coded using the Iowa Drug Information System (IDIS) ingredient codes (27). Participants provided informed consent after the study’s approval by the institutional review boards of the University of Pittsburgh and the University of Tennessee, Memphis.
Of the 840 women, 12 did not have CT scans available, 18 did not have summed grip strength measures, 5 did not have a physical function summary score, and 96 women did not undergo knee extensor strength testing. Stringent eligibility criteria for the knee extensor strength test implemented in Health ABC; that is, elevated blood pressure, previous stroke, severe bilateral knee pain, or bilateral knee replacement were the reasons for exclusion. As a result, the subject number varies for analyses of these outcome variables resulting in 828 participants for CT analyses, 822 for grip strength, 744 for knee extensor strength, and 835 for physical function.
Axial scans of the midthigh were obtained by computer tomography (Pittsburgh site: GE CT-9800 Advantage, General Electric, Milwaukee, WI; Memphis site: Somatom Plus 4, Siemens, Erlangen, Germany or PQ 2000S, Marconi Medical Systems, Cleveland, OH). Subjects were in the supine position with their legs flat on the table. The measurement used in this analysis was the leg used in isokinetic knee extensor strength testing, which was generally the right. To locate the midthigh scan position, an anterior-posterior scout of the entire femur was obtained. The femoral length was measured in cranial-caudal dimension and the scan position determined as midway between the medial edge of the greater trochanter and the intercondyloid fossa and perpendicular to the shaft of the femur. A single axial image of 10-mm slice thickness was obtained (scanning parameters were 120 kVp and 200–250 mAs). The effective whole body radiation dose was 15 μSv for the anterior-posterior thigh scout scan and 30 μSv for the axial slice. Midthigh quadriceps and hamstring muscles, subcutaneous and intermuscular fat areas were derived from the CT scan using software written at the University of Colorado Health Sciences Center (CT scan Reading Center for Health ABC) using IDL development software (RSI Systems, Boulder, CO). A quality review was undertaken on all images acquired to ensure appropriate quality. This involved determination that the images were present, that the proper scan techniques were used (femur length, scan location, slice thickness, kVp, mAs), and that pathology and artifacts in the image were absent. Soft-tissue type was determined using a bimodal image histogram resulting from the distribution of CT numbers in adipose tissue and muscle tissue. Manual drawing of a line along the deep fascial plane surrounding the muscles separated the intermuscular from subcutaneous fat. Individual muscles were identified and manual tracing separated the quadriceps and hamstring muscles. Areas were calculated by multiplying the number of pixels of a given tissue type by the pixel area. Skeletal muscle attenuation for the quadriceps and hamstring muscles was measured as the mean attenuation coefficient in Hounsfield units (HU). Muscle attenuation has been used in a number of studies to represent muscle tissue composition and quality (15,22). Reproducibility of areas and attenuation was assessed by random reanalysis of 5% of the images and was within 5%.
Dynamic knee extensor strength was measured at an angular velocity of 60°·s−1 on a Kin-Com 125 AP isokinetic dynamometer (Chattanooga, TN). Participants were familiarized to the testing procedure and then completed a maximum of 6 trials. Start and stop angles were set at 90 and 30°, and from the first attempt, the torque produced over the entire range was plotted, with the plot of each subsequent attempt overlaid on the previous trials until three similar curves were obtained, with the average maximum torque determined from these three trials. The coefficient of variation (repeated testing of 63 participants) for isokinetic knee extensor strength was 10.7%. Isometric hand grip strength (right and left) was determined in duplicate by a handheld dynamometer with the maximum summed value reported (Jamar; TEC, Clifton, NJ). The reported CV for hand grip strength is less than 2.7% (23).
Physical function was determined by the Health ABC physical function summary scale, which has been described in detail previously (32). Briefly, the scale is an extension of the lower-extremity performance tests used in the Epidemiologic Studies of the Elderly (EPESE) (18) and includes five repeated chair stands, usual 6-m walk time, a 6-m narrow walk between lines 20 cm apart, and a balance test (30-s semi- and full-tandem stands and a 30-s single-leg stand). A ratio score from 0 to 1 was created for each test and the four tests added to provide a continuous scale from 0 to 4. In developing the ratio score, repeated chair stand time was converted to chair stands per second, 6-m walk times to meters per second, and time for the balance tests summed to give a score out of 90 s. When the test was not successfully undertaken, 0 was given. Then, based on performance data derived from older adults (32), each score was divided by the possible maximal performance so that chair stands was divided by 1, 6-m walks were divided by 2 m·s−1, and standing balance was divided by 90 s to give a score from 0 to 1. The Health ABC summary scale was developed to discriminate among older persons with a broad range of functional ability.
Height and weight were obtained using a Harpenden stadiometer (Holtain, Wales, UK) and a standard balance beam scale, respectively. Body mass index (BMI) was calculated as weight divided by square height (kg·m−2). Bone mineral–free lean mass (LM) and fat mass (FM) were assessed by dual x-ray absorptiometry (DXA; Hologic 4500A, Waltham, MA, version 8.21a). Smoking history (never, former, current), health history, education, physical activity, weight training in the past year, and medication usage were determined by an interviewer-administered questionnaire. Years of education was coded as less than high school, high school graduate, and post secondary. Self-rated health was on a five-point scale from poor to excellent. Chronic conditions assessed were diabetes, arthritis, pulmonary disease, coronary heart disease (CHD), stroke, and peripheral vascular disease. Pulmonary disease (prevalent obstructive/restrictive) was determined from a pulmonary function test and included mild, moderate, and severe obstructive or restrictive pattern as well as normal function after bronchodilator usage. Physical activity and exercise were ascertained by a standardized instrument derived from the Leisure Time Physical Activity Questionnaire (36). Five activity levels were calculated based on kilocalories per week expended on walking, stair climbing, and moderate and vigorous physical activity. The levels were: 1) <500 kcal·wk−1; 2) ≥500 kcal·wk−1 but <1000 kcal·wk−1; 3) ≥1000 kcal·wk−1 but <1500 kcal·wk−1; 4) ≥1500 kcal·wk−1 but <2000 kcal·wk−1; and 5) ≥2000 kcal·wk−1. Participation in weight training was determined from the question, “In the past 12 months, did you do any weight or circuit training, at least 10 times?” Fifty-five non-ERT women and 34 ERT users responded in the affirmative to this question. Oral and inhaled corticosteroid, and angiotensin-converting enzyme (ACE) inhibitor usage, which have direct and indirect effects on muscle tissue (26), was determined from drug data coded using IDIS ingredient codes (27). In addition to the potential confounders, total cholesterol and high-density lipoprotein ((HDL) cholesterol were measured by a calorimetric technique (Johnson &Johnson Vitros 950 analyzer, New Brunswick, NJ), with low-density lipoprotein (LDL) cholesterol calculated using the Friedewald equation.
Data were analyzed using the SPSS (version 11.5, SPSS Inc., Chicago, IL) statistical software package and included descriptive statistics, t-tests, χ2 analysis, and ANCOVA. Multivariate analyses using ANCOVA to examine the association of ERT usage with muscle composition, strength, and physical function were adjusted for potential confounders, which included in the first model, age, height, weight, clinic site, and education. In the second set of models, we additionally adjusted for self-reported health status, smoking, diabetes, arthritis, pulmonary disease, CHD, stroke, and peripheral vascular disease. In the final models, we also adjusted for weight training in the past year, physical activity, and corticosteroid and ACE inhibitor usage. Where appropriate, the Bonferroni post hoc procedure for multiple comparisons was employed to locate the source of significant differences in means. As there were minimal differences among the three models in the analyses, only the final fully adjusted models are shown. In addition to comparing current estrogen users and nonusers, subanalyses were performed comparing never-users with current users, and never-users, current users, and HRT users. All tests were two-tailed and statistical significance set at P < 0.05. Results are given as the mean ± SD, unless stated otherwise.
Subject characteristics are shown in Table 1 for current users and nonusers of ERT. Current users commenced taking estrogen at 53.7 ± 11.2 yr of age for an average of 17.8 ± 10.8 yr. When characterized by quartiles and adjusted for potential confounding factors, length of ERT usage was not associated with any outcome measure. Women not currently taking ERT were slightly older than women using ERT and also heavier. Education level was different between the two groups, with proportionally more ERT users having undertaken postsecondary studies, although a similar percentage were high school graduates and above. There was no difference between groups for physical activity. Regarding chronic conditions, the only difference by ERT status was for pulmonary disease, with a higher level in ERT users. In addition, as expected with estrogen use, ERT users had an improved lipid profile with higher HDL (66 ± 18 vs 57 ± 15 mg·dL−1, P < 0.001) and lower LDL (104 ± 29 vs 130 ± 35 mg·dL−1, P < 0.001) and total cholesterol (201 ± 32 vs 217 ± 39 mg·dL−1, P < 0.001). Of the 581 women not currently taking ERT, 133 were former users, with an average use of 8.9 ± 9.3 yr; however, their lipid values were the same as never-users (HDL, 56 ± 15 vs 57 ± 15 mg·dL−1, P = 0.868), LDL (130 ± 35 vs 129 ± 33 mg·dL−1, P = 0.767), and total cholesterol (218 ± 39 vs 217 ± 36 mg·dL−1, P = 0.904).
Muscle attenuation and cross-sectional area were higher (P < 0.05) in ERT users than nonusers for the quadriceps muscles, with no difference for the hamstring muscles (Table 2). The differences for the quadriceps muscles were modest, with the difference in attenuation and CSA being 2.5 and 3.3%, respectively. There were no differences between groups in subcutaneous or intermuscular fat CSA. Hand grip strength was also higher in estrogen users (by 3.2%), whereas knee extensor strength approached significance (difference between groups of 3.1%; P < 0.10). However, when knee extensor torque was divided by quadriceps CSA to calculate specific torque, there was no difference by ERT usage. Similarly, lower-extremity physical performance was not different between groups. The results were unchanged when BMI was substituted for height and weight, and lean and fat mass were substituted for weight in the multivariate models.
Subanalysis comparing never-users with current users of estrogen replacement is shown in Table 3. The results were unchanged from those in the entire group comparing current users and nonusers, with quadriceps attenuation and CSA as well as hand grip strength being greater in ERT users (with the same magnitude of difference), with no difference for any other outcome variable. Substitution of BMI for height and weight, and lean and fat mass for weight, did not alter the models.
Of the current ERT users, 79 were also taking progestogens in combination with estrogen (hormone replacement therapy, HRT). Consequently, we compared current ERT users and HRT users to nonusers to determine whether the combined hormone treatment exerted different effects than estrogen alone (Table 4). A difference remained for quadriceps CSA, with muscle size being greater in ERT than non-ERT users. The smaller subject number for HRT users with CT data (N = 77) and greater variability resulted in no statistical difference between this group and ERT and non-ERT users. There was also no difference between ERT and HRT use for quadriceps HU and grip strength, with differences among the three groups approaching significance. The results were unchanged when BMI was included instead of height and weight in the analyses. When fat and lean mass were included instead of body weight, subcutaneous fat CSA of HRT users was greater (P = 0.003) than non-ERT and ERT users.
Since the findings of the Womens’ Health Initiative were reported (3,30), the efficacy of estrogen treatments in postmenopausal women has been questioned (21). In weighing up the benefits and risks of estrogen use, it would seem appropriate that skeletal muscle should also be considered as a target tissue because of its importance in the performance of daily activities. Given that many older people, especially women, may perform activities close to their maximum capacity, small changes in muscle function may be the difference between independence and dependence. The results of the present study, in a large cohort of older well-functioning women who underwent CT scanning, indicates that estrogen replacement is associated with minor enhancement in some components of muscle composition and strength compared with nonusers, although this did not translate into improved physical function.
Of the muscle composition parameters examined in this study, only the CSA and attenuation (radiological density) for the quadriceps muscles were greater in estrogen users than nonusers, although the magnitude of difference was small and less than that for the reproducibility of the technique. The difference by muscle location may reflect an interaction between muscle usage and estrogen replacement. In activities of daily living, such as rising from a chair, walking, and stair climbing, the knee extensor muscles are subject to greater recruitment than the knee flexors (37), and as a result may be more responsive to actions of estrogen. The lower functional demands of the hamstring muscles may also be responsible for the higher fat infiltration, as determined by the lower HU, compared with the quadriceps muscles.
Cross-sectional reports indicate that estrogen deprivation associated with menopause leads to a decline in lean tissue mass (1). In a recent randomized placebo-controlled trial of women within 5 yr of the menopause, Sipilä et al. (33) found that 12 months of HRT treatment resulted in a 6% increase in quadriceps muscle CSA. Hormone treatment in the study by Sipilä et al. (33) was estradiol and norethisterone acetate administered continuously, and the anabolic effect may have been derived from the synthetic progesterone, which was derived from testosterone. Similarly, others (13) have reported short- and long-term HRT administration to enhance whole body lean mass in postmenopausal women. In our cohort, we examined whether HRT users had larger muscle CSA compared with ERT users to tease out the additional effects of progestogens; however, there was no difference between the two groups.
Estrogen administration enhances growth hormone (GH) release in postmenopausal women (12), which may have an anabolic effect on muscle tissue. In addition, estrogen receptors are present in human skeletal muscle (24), so it is possible that a direct effect may also be derived. However, in contrast to the above studies indicating a positive effect of estrogen on lean tissue, there are cross-sectional reports of no difference in lean tissue mass by DXA and muscle CSA by ultrasound between users and nonusers of ERT (6,16). Likewise, in experimental trials where estrogen (8) or HRT (2,9) were administered, or where exercise training was conducted in users of ERT and HRT (11), hormone use did not influence lean mass change.
However, CSA of the muscle does not indicate the quality of the tissue, as CSA may be increased by fatty infiltration. Animal (19) and human studies (10) indicate that estrogen promotes lipolysis, enhancing fatty acid availability and oxidation, and sparing muscle and liver glycogen. In addition to the metabolic effect, the lypolytic effects of GH may also influence muscle lipid content and fatty infiltration. We found a minor difference (2.5%) between users and nonusers of ERT for muscle attenuation of the quadriceps, with no difference for the hamstring muscles. To date, no other studies have reported the effects of estrogen use in older postmenopausal women on muscle density, and therefore comparisons with other studies examining the effects of estrogen on attenuation are not possible. In a trial of HRT in postmenopausal women for 6–12 months, Skelton and colleagues (34) found an increase in strength of the adductor pollicis muscle occurred without an increase in muscle CSA. It is possible that muscle density or quality may have improved with hormone treatment, and that this contributed to the increase in strength without an accompanying increase in CSA.
A statistically significant, yet minor, difference in grip strength existed between ERT users and nonusers in the present study, whereas that for knee extensor strength approached significance. The percent difference in strength by estrogen status was similar to that for quadriceps muscle CSA and density. Phillips and colleagues (28) previously reported that hormone replacement preserved specific force of the adductor pollicus muscle following menopause. Moreover, several recent experimental trials indicate that estrogen or HRT preserves or enhances muscle performance in postmenopausal women, with differences of up to 13% between hormone users and nonusers for various muscle groups (17,20,33,34).
In contrast, several cross-sectional studies that examined hand grip as well as the strength of other upper- and lower-limb muscle groups found no beneficial effect of estrogen therapy (6,31,35). In a similar fashion, randomized controlled trials by Armstrong et al. (4) and Ribom and coworkers (29) reported that estrogen or HRT did not influence muscle performance. It should be noted that the relatively large number of users and nonusers in our study resulted in statistically significant differences in grip strength, which were no greater than those found in other studies that concluded no difference based on ERT status. Moreover, in contrast to Phillips et al. (28), who reported specific force to increase with hormone treatment, specific torque of the quadriceps was not different in our group of estrogen users and nonusers, suggesting that the small statistically significant difference in muscle composition between users and nonusers has little physiological relevance.
The critical question is, if estrogen use is associated with improved muscle composition and strength, does this translate into improved physical function? Compared with muscle strength, only scant information exists on the effect of estrogen replacement on aspects of physical function. Seeley et al. (31) reported similar gait speed, chair rise ability, and balance between users and nonusers of estrogen replacement in the Study of Osteoporotic Fractures. Similarly, Armstrong et al. (4) and Brooke-Wavell et al. (7) found no significant effect by hormone use on postural stability, although Naessen and colleagues (25) reported postural balance function to be enhanced in long-term estrogen users compared with nonusers. In the present study, we used a summary scale that captured different dimensions of physical function and found no effect of estrogen use (or combined estrogen-progestogens) on performance. The lack of effect on physical function in our study is not surprising given the magnitude of difference between users and nonusers for quadriceps muscle CSA, density, and strength. In addition, our subjects were well functioning, and subtle differences in muscle composition or strength may have no measurable effect on physical performance tasks.
Given that our cohort are of well-functioning white women aged 70–79 yr, they are not representative of all postmenopausal women, and indeed age and years since menopause, and type and length of hormone usage may partly explain the variation found in cross-sectional and experimental studies reported above. Moreover, our study was cross-sectional, and as such we cannot infer cause and effect. However, we examined a large number of users and nonusers of ERT who had CT scans performed, and as such we were able to examine muscle attenuation in addition to muscle CSA. Further, we were able to examine strength of both the upper and lower extremity as well as physical function, and these analyses were adjusted for a number of potential confounders.
In conclusion, when combined with reports in the literature, it appears that estrogen replacement has at best a minor association with muscle composition and strength, and these associations in well-functioning older women are of questionable clinical relevance. However, it is important to note that skeletal muscle benefits derived from resistance exercise far outweigh any effect likely achieved with estrogen replacement, and should be promoted actively if muscle and physical performance benefits are desired.
1. Aloia, J. F., D. M. McGowan, A.N. Vaswani, P. Ross, and S. H. Cohn. Relationship of menopause to skeletal and musclemass. Am. J. Clin. Nutr.
2. Aloia, J. F., A. Vaswani, L. Russo, M. Sheehan, and E. Flaster. The influence of menopause and hormonal replacement therapy on body cell mass and body fat mass. Am. J. Obstet. Gynecol.
3. Anderson, G. L., M. Limacher, A. R. Assaf, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women’s Health Initiative randomized controlled trial. JAMA
4. Armstrong, A. L., J. Osborne, C. A. C. Coupland, M. B. Macpherson, E. J. Bassey, and W. A. Wallace. Effects of hormone replacement therapy on muscle performance and balance. Clin. Sci.
5. Belchetz, P. E. Hormonal treatment of postmenopausal women. N. Engl. J. Med.
6. Bemben, D. A., and D. B. Langdon. Relationship between estrogen use and musculoskeletal function in postmenopausal women. Maturitas
7. Brooke-Wavell, K., G. M. Prelevic, C. Bakridan, and J. Ginsburg. Effects of physical activity and menopausal hormone replacement therapy on postural stability in postmenopausal women – a cross-sectional study. Maturitas
8. Davis, S. R., K. Z. Walker, and B. J. Strauss. Effects of estradiol with and without testosterone on body composition and relationships with lipids in postmenopausal women. Menopause
9. D’Eon, T. and B. Braun. The roles of estrogen and progesterone in regulating carbohydrate and fat utilization at rest and during exercise. J. Womens Health Gend. Based Med.
10. D’Eon, T. C. Sharoff, S. R. Chipkin, D. Grow, B. C. Ruby, and B. Braun. Regulation of carbohydrate metabolism by estrogen and progesterone in women. Am. J. Physiol. Endocrinol. Metab.
11. Figueroa, A., S. B. Going, L. A. Milliken, et al. Effects of exercise training and hormone replacement therapy on lean and fat mass in postmenopausal women. J. Gerontol. Med. Sci.
12. Friend, K. E., M. L. Hartman, S. S. Pezzoli, J. L. Clasey, and M. O. Thorner. Both oral and transdermal estrogen increase growth hormone release in postmenopausal women–a clinical research center study. J. Clin. Endocrinol. Metab.
13. Gambacciani, M., M. Ciaponi, B. Cappagli, L. De Simone, R. Orlandi, and A. R. Genazzani. Prospective evaluation of body weight and body fat distribution in early postmenopausal women with and without hormonal replacement therapy. Maturitas
14. Goodpaster, B. H., D. E. Kelley, F. L. Thaete, J. He, and R. Ross. Skeletal muscle attenuation determined by computed tomography
is associated with skeletal muscle lipid content. J. Appl. Physiol.
15. Goodpaster, B. H., F. L. Thaete, J. A. Simoneau, and D. E. Kelley. Subcutaneous abdominal fat and thigh muscle composition predict insulin sensitivity independently of visceral fat. Diabetes
16. Gower, B. A., and L. Nyman. Associations among oral estrogen use, free testosterone concentration, and lean body mass among postmenopausal women. J. Clin. Endocrinol. Metab.
17. Greeves, J. P., N. T. Cable, T. Reilly, and C. Kingsland. Changes in muscle strength
in women following the menopause: a longitudinal assessment of the efficacy of hormone replacement therapy. Clin. Sci.
18. Guralnick, J. M., E. M. Simonsick, L. Ferrucci, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J. Gerontol. Med. Sci.
19. Hansen, F. M., N. Fahmy, and J. H. Nielsen. The influence of sexual hormones on lipogenesis and lipolysis in rat fat cells. Acta Endocrinol.
20. Heikkinen, J., E. Kyllönen, E. Kurttila-Matero, et al. HRT and exercise: effects on bone density, muscle strength
and lipid metabolism. A placebo controlled 2-year prospective trial on two estrogen-progestin regimens in healthy postmenopausal women. Maturitas
21. Hulley, S. B., and D. Grady. The WHI estrogen-alone trial – do things look any better? JAMA
22. Kelley, D. E., B. S. Slasky, and J. Janosky. Skeletal muscle density
: effects of obesity and non-insulin-dependent diabetes mellitus. Am. J. Clin. Nutr.
23. Kritz-Silverstein, D., and E. Barrett-Connor. Grip strength and bone mineral density in older women. J. Bone Miner. Res.
24. Lemoine, S., P. Granier, C. Tiffoche, F. Rannou-Bekono, M-L. Thieulant, and P. Delamarche. Estrogen receptor alpha mRNA in human skeletal muscles. Med. Sci. Sports Exerc.
25. Naessén, T., B. Lindmark, and H. C. Larsen. Better postural balance in elderly women receiving estrogens. Am. J. Obstet. Gynecol.
26. Onder, G, B., J. W. H. Penninx, R. Balkrishnan, et al. Relation between use of angiotension-converting enzyme inhibitors and muscle strength
and physical function in older women: an observational study. Lancet
27. Pahor, M., E. A. Chrischilles, J. M. Guralnik, S. L. Brown, R. B. Wallace, and P. Carbonin. Drug data coding and analysis in epidemiologic studies. Eur. J. Epidemiol.
28. Phillips, S. K., K. M. Rook, N. C. Siddle, S. A. Bruce, and R. C. Woledge. Muscle weakness in women occurs at an earlier age than in men, but strength is preserved by hormone replacement therapy. Clin. Sci.
29. Ribom, E. L., K. Piehl-Aulin, S. Ljunghall, Ö. Ljunggren, and T. Naessén. Six months of hormone replacement therapy does not influence muscle strength
in postmenopausal women. Maturitas
30. Rossouw, J. E., G. L. Anderson, R. L. Prentice, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA
31. Seeley, D. G., J. A. Cauley, D. Grady, W. S. Browner, M. C. Nevitt, and S. R. Cummings. Is postmenopausal estrogen therapy associated with neuromuscular function or falling in elderly women? Arch. Intern. Med.
32. Simonsick, E. M., A. B. Newman, M. C. Nevitt, et al. Measuring higher level physical function in well-functioning older adults: Expanding familiar approaches in the health ABC Study. J. Gerontol. Med. Sci.
33. Sipilä, S, D. R., Taaffe, S. Cheng, J. Puolakka, J. Toivanen, and H. Suominen. Effects of hormone replacement therapy and high-impact physical exercise on skeletal muscle in post-menopausal women: a randomized placebo-controlled trial. Clin. Sci.
34. Skelton, D. A., S. K. Phillips, S. A. Bruce, C. H. Naylor, and R. C. Woledge. Hormone replacement therapy increases isometric strength of adductor pollicis in post-menopausal women. Clin. Sci.
35. Taaffe, D. R., M. L. Villa, R. Delay, and R. Marcus. Maximal muscle strength
of elderly women is not influenced by oestrogen status. Age Aging
36. Taylor, H. L., D. R. Jacobs, Jr. B., Schucker, J. Knudsen, A. S. Leon, and G. Debacker. A questionnaire for assessment of leisure time physical activities. J. Chron. Dis.
37. Wretenberg, P., and U. P. Arborelius. Power and work produced in different leg muscle groups when rising from a chair. Eur. J. Appl. Physiol.