| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Department of Health and Environment, Division of Obstetrics and Gynecology, Division of Otorhinolaryngology, Faculty of Health Sciences, University Hospital of Linköping, Linköping; and the Department of Obstetrics and Gynecology, County Hospital of Jönköping, Jönköping, Sweden.
Address reprint requests to: Sara Ekblad, MD Division of Obstetrics and Gynecology Department of Health and Environment Faculty of Health Sciences University Hospital Linköping S-581-85 Sweden E-mail: matha{at}gyn.liu.se
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: Forty healthy postmenopausal women without vasomotor symptoms were randomized to transdermal 17ß-estradiol (E2) 50 µg/day for 14 weeks or identical transdermal placebo patches. Postural balance was measured with dynamic posturography before and after 4, 12, and 14 weeks of therapy. In this test, the visual, vestibular, and somatosensory systems were provoked with increasing difficulty and body sway was measured with a dual forceplate. A low score showed large sway and a score of 100 showed no sway at all.
Results: Thirty-eight women completed the study. Both groups had normal balance for their ages and near maximum scores in the three easier balance tests at baseline. In the most difficult test, both groups improved their postural balance significantly (from 13 to 32 and from 22 to 39, respectively) after 4 weeks. Thereafter, no change was seen. One problem was low statistical power, but the relative change in balance did not differ between groups. The comparison did not show even a minute advantage of E2 over placebo, so a study with higher power would probably not have shown a more pronounced effect of estrogen than placebo. The change over time did not differ between groups, which indicates a significant learning effect.
Conclusion: In women without vasomotor symptoms, estrogen therapy did not seem to increase postural balance significantly more than placebo. However, we could not rule out that estrogens affect postural balance in women with vasomotor symptoms.
It is believed that estrogens might prevent fractures by decreasing bone loss and stimulating postural balance,1–4 decreasing the risk of falling. Postural balance is a delicate function that depends on the integration of several different systems. The vestibular apparatus registers the position of the head and its dynamic changes; the visual system integrates its signals with those from the vestibular system. The proprioceptive or somatosensory system registers small changes in tension within muscles and tendons. Integration among those three systems in the cerebellum and brainstem is necessary for well-functioning balance and minimal risk of falling.5,6 Dynamic posturography has been useful for evaluating components of balance and their integration quantitatively and qualitatively.5,6
In a previous open study, we found a significant improvement in postural balance measured with dynamic posturography in 19 postmenopausal women with vasomotor symptoms, after 4 weeks of transdermal estradiol (E2) therapy.4 The most evident increases were in the most difficult tests, when a perfect integration of afferent signals within the central nervous system (CNS) is required. That observation might have been due to a generally stimulatory effect on the CNS or the fact that those women had been relieved of their vasomotor symptoms, which previously affected their well-being, sleep, and balance. In the latter case, balance would not improve after estrogen therapy given to postmenopausal women who had no vasomotor symptoms.
Repeated measurements with dynamic posturography might improve balance through a learning process, a theory that merits a randomized placebo-controlled trial of estrogens and effects on balance. A study on effects of estrogens on women who had no vasomotor symptoms is also important because those women have not suffered from disturbances in sleep and well-being caused by flushes. The aim of this study was to assess whether transdermal estrogen improved balance more than placebo in postmenopausal women who had no vasomotor symptoms.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
error of .05 and a ß error of .20 (corresponding to a power of .8). Such a sample was considered impossible to recruit in a clinical setting, especially when taking into account that we intended to study women totally without vasomotor symptoms, volunteering to take part in the study. The women without vasomotor symptoms were recruited using data from a previous questionnaire study sent to all 1324 women who were 55 or 56 years old in Linköping, Sweden.7 The questionnaire asked about months since menopause, vasomotor symptoms, use of hormonal replacement therapy (HRT), gynecologic operations, and so forth. Women whose responses indicated they met inclusion criteria for the present study, ie, who were spontaneously at least 6 months postmenopausal and had reported never having had vasomotor symptoms, and who did not use HRT, were invited by letter to participate in the present study. About 75% of invited women were willing to participate and came to an inclusion visit.
At the first visit, the inclusion criteria again were checked. Exclusion criteria were histories of vasomotor symptoms, rotatory vertigo, or unsteadiness. Women with malignant diseases, hepatic or thromboembolic diseases, neurologic diseases, or other serious illnesses, and women who used drugs that could interfere with postural balance were also excluded. No women were taking chronic medication, except for four who were taking antihypertensive drugs and one who was taking thyroxin. Those women were stable in their diseases, and therapies were unchanged throughout the study period. All women received oral and written information on the study and gave oral consent. Enrollment was from February 1996 to February 1997.
At inclusion, women had gynecologic, otolaryngologic, and otoneurologic examinations and mammography. Blood pressure (BP) was measured sitting and standing. Arm strength was tested by hydraulic hand dynamometer.8 Each woman was encouraged to make three efforts with each hand. The mean values of dominant and nondominant arms were registered. Blood samples were drawn for analysis of serum concentrations of FSH and E2 by radioimmunoassay methods at the Department of Clinical Chemistry at the University Hospital of Linköping, Sweden.
Forty women were randomized into two groups of 20 each who received identical transdermal patches with 17ß-E2 50 µg/day or placebo for 14 weeks. The patches were prepackaged and coded by the company, which also randomized the subjects, who were included consecutively. Randomization was generated at the Biostatistical Department of Ciba Geigy AB, Gothenburg, Sweden, using a computerized standard random table model. During the study the codes were kept in a sealed envelope in case of emergency, but no code was broken before the study was completed. Doctors, nurses, and subjects were masked to assignments. During the 13th and 14th treatment weeks, all women were given 10 mg oral medroxyprogesterone acetate per day to induce withdrawal bleeding in those who had received E2.
Climacteric symptoms were assessed with a Kupperman index9 before the study, after 4 and 12 weeks of E2 or placebo, and during the 14th week, when medroxyprogesterone acetate was added. Kupperman index measures the most common menopausal complaints. The symptoms are converted into figures based on severity grade (graded 0 to 3). The severity scores for sweating, sleep disturbances, and nervousness are multiplied by 2 and the score for hot flushes by 4. The highest possible score is 51 and the classical climacteric symptoms hot flushes and sweating are given the highest weight. The women also reported any bleeding, vasomotor symptoms, or other adverse effects in a diary beginning 1 week before treatment and throughout the study. The diary was brought to the register nurse at each clinical visit.
Each subject had dynamic posturography before treatment, after 4 and 12 weeks, and during the 14th week. The aim of dynamic posturography is to assess and quantify body sway objectively. A normal body sway has two main strategies, one of which is ankle sway, when the body acts as an inverted pendulum. Ankle sway is most common when movements are small. If the degree of sway is increased, a new sway pattern will develop: the hip strategy. By using the hips, the degree of sway can increase, and stability is maintained by sheer force. Dynamic posturography was done using Equitest version 4.04 (Neurocom International Inc., Clackamas, OR). The method has been thoroughly described as a reliable way to evaluate balance.5,6,10 The subject stands enclosed by a visual background with feet parallel on a dual forceplate. A safety harness keeps the subject from falling. The dual forceplate measures the center of gravity by strain gauges in each corner, records vertical forces between feet and ground, and horizontal anterior-posterior forces. The maximum anterior-posterior sway range without falling is approximately 12°. By estimating sway degree, a continuous score from 0 to 100 is created. A score of 100 is equivalent to a sway of 0°; thus, a very low score denotes large sway, and a score of 0 is a sway of more than 12° or a fall. The forceplate and the visual background can be moved with the anterior-posterior sway of the subject (sway referenced), which causes a sensory conflict between visual, vestibular, and somatosensory systems, creating problems maintaining balance.
Dynamic posturography is composed of two main tests, sensory organization and movement coordination. Movement coordination was done on all women only at study entry to exclude dysfunction of afferent and efferent somatosensory system as seen in polyneuropathies. Sensory organization is divided into six separate tests, each lasting 20 seconds, with tests 4, 5, and 6 repeated three times, as seen in Figure 2 of Hammar et al.4 In the first and second tests, the subject stands with eyes open or closed on a stable platform. In the third test, the subject’s eyes are open on a stable platform, with sway-referenced visual background. In the fourth and fifth tests, the subject’s eyes are open and closed, respectively, on a sway-referenced platform, with stable visual background. In the sixth test, the subject’s eyes are open on a sway-referenced platform, with sway-referenced visual background.4 The most difficult tests are the fifth and sixth because they cancel visual and distort somatosensory inputs, or give distorted information from visual and somatosensory systems. The precision error was below 5% for the three most difficult sensory orientation tests.11
This study included all the tests described in our previous study,4 which included some more difficult and challenging tests than commonly used in women with balance disturbances. The reason for increasing test difficulty was that dynamic posturography mainly tests subjects with balance problems and vertigo. The apparatus allows increase of the gain of movements of the platform and the visual background. For example, when the sway of the subject is 4° forward, the visual background and the platform can move 6° (gain 1.50) or nearly 8° (gain 1.99), which increases the difficulty of the test and provokes unsteadiness. We used gain 1.99 throughout the study for the 38 women without vasomotor symptoms who completed the study.
All data are presented as median values and 25th to 75th percentiles or mean ± standard deviation (SD), depending on whether normality was checked and verified. Comparisons were made with the nonparametric Mann-Whitney U test. To compare results from dynamic posturography in different groups, we used analyses of variance with repeated measures after testing that results in each group were normally distributed, according to the Kolmogrov-Smirnov test. The study was approved by the local ethics committee of the Faculty of Health Sciences, University of Linköping.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). The gynecologic and otoneurologic examinations at inclusion were normal, and no abnormalities were found by mammography. During the study, none of the women developed disease or changed medications that affect postural balance. Two women in the placebo group left the study, one owing to local skin reaction and one to psychiatric problems. The other 38 women completed all tests. No women developed climacteric symptoms12 throughout the study or 3 months after the study according to Kupperman index and diaries.
|
). Serum E2 concentrations increased and FSH decreased significantly (P < .001) in women treated with 17ß-E2 but not in the placebo group (Table 2), indicating that the E2 group was compliant. Blood pressure decreased slightly but stayed within normal limits from the first to the last measurement in both groups (Table 2). Arm strength did not change in any group, whether dominant or nondominant arm. Kupperman index did not change significantly after E2 or placebo.
|
. The results of the three easier balance tests are not shown in Table 3 because they were near maximum at baseline and did not change significantly with time. The data from the dynamic posturography did not deviate from the normal distribution according to the Kolmogrov-Smirnov test; therefore, parametric analysis of variance was used for multiple comparisons in this study. Using analysis of variance with repeated measures for the factors time (four different occasions) and tests (six different tests), no significant effects of treatment (versus placebo) were found. Also, when we applied analysis of variance with repeated (time) measures for each sensory organization test condition separately, no significant effects of treatment regimens were unveiled. All women in the double-masked test increased their balance to some extent. The time and test factors were significant, suggesting a significant learning effect which was different across tests. The learning effect was most prominent in the three most difficult tests. The multiple comparisons showed that both groups increased their balance significantly with time, whereas the change did not differ between estrogen- and placebo-treated women.
|
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
In our previous study, women with hot flushes increased their postural balance significantly after 4 weeks of transdermal therapy with 17ß-E2.4 To determinate whether that improvement was due to something more than a learning effect, we used analyses of variance repeated measure to compare our new results with those from the previous study.4 In that study, women were included with vasomotor complaints, received identical transdermal E2 treatment, and had the same dynamic posturography tests at the same intervals as the women in the present study. We found an increase in balance during estrogen therapy in the most difficult test (sensory organization test 6), which was significantly higher (P < .045) in women with vasomotor symptoms4 than in women from the present study without symptoms, indicating that a test-retest factor might not have been the only reason for improved balance in women with vasomotor symptoms. Thus, estrogens might directly affect postural balance by increasing well-being and sleep in women with vasomotor symptoms. However, that interpretation is jeopardized by the fact that we compared data from two studies, one of which was open and the other masked. We also used the highest gain in all women in the present study, whereas only 14 of 19 women in the former study were tested with the highest gain, which might suggest that some women with vasomotor symptoms had slight subclinical disturbances in their balance.
It could be argued that the statistical power of the present study was too low to show a difference between groups. We chose a sample of 20 women in each group presuming that we would not see a learning effect in either. In such case, the placebo group would not have changed their postural balance throughout the study and the amount of change in the estrogen group would have been about the same as in our previous study.4 We observed clear learning effects over time in both groups that corrupted that assumption. The relative change in balance did not differ between groups. The comparison did not show even a minute advantage from E2 over placebo, so a study with higher power would probably not have shown a more pronounced effect of estrogen over placebo, even if we had recruited the number of subjects suggested by the power calculation. No increase in arm strength was seen in either group during estrogen therapy, which agrees with Armstrong et al16 and Brown et al.17
Naessen and co-workers18 found better postural balance in 11 postmenopausal long-term users of 17ß-E2 than 16 age-matched nonusers, measured with a computerized sway platform test. That agreed with our previous study,4 although different methods of measuring balance were used. Goebel et al19 found no significant increase in balance after 8 months of estrogen replacement therapy, measured with repeated dynamic posturography. Neither Seeley and co-workers20 nor Armstrong et al16 found beneficial effects of estrogen therapy on muscle strength, postural balance, or risk of falling. In those studies, women with or without vasomotor symptoms were not analyzed separately. Our recent data suggest that only women with vasomotor symptoms increase their balance after estrogen treatment, so the studies are not fully comparable. Our study population probably differed from the general population because of selection methods, recruiting only women without vasomotor symptoms. It is therefore essential to make a double-blind placebo-controlled study of effects of E2 on balance in postmenopausal women with vasomotor symptoms.
|
Footnotes |
|---|
Financial support was provided by Trygg-Hansa Research Foundation (Stockholm, Sweden), the Swedish Medical Research Council (Stockholm, Sweden) project no. K98-17X-12651-01A, and the study drugs and placebo were provided by Novartis (Stockholm, Sweden).
Received February 1, 1999. Received in revised form July 21, 1999. Accepted August 5, 1999.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Hayes WC, Piazza SJ, Zysset PK. Biomechanics of fracture risk prediction of the hip and spine by quantitative computed tomography. Radiol Clin North Am 1991;29:2–18.
3. Cummings SR, Nevitt MC, Browner WS, Stone K, Fox K, Ensrud K, et al. Risk factors for hip fracture in white women. N Engl J Med 1995;332:767–73.
4. Hammar ML, Lindgren R, Berg GE, Möller CG, Niklasson MK. Effects of hormonal replacement therapy on the postural balance among postmenopausal women. Obstet Gynecol 1996;88:955–60.[Abstract]
5. Nashner LM, Peters JF. Dynamic posturography in the diagnosis and management of dizziness and balance disorders. Neurol Clinics 1990;8:331–49.[Medline]
6. Voorhees RL. The role of dynamic posturography in neurootologic diagnosis. Laryngoscope 1989;99:995–1001.[Medline]
7. Ivarsson T, Spetz A-C, Hammar M. Physical exercise and vasomotor symptoms in postmenopausal women. Maturitas 1998;29:139–46.[Medline]
8. Mathiowetz V, Kashman N, Volland G, Weber K, Dowe M, Rogers S. Grip and pinch strength: Normative data for adults. Arch Phys Med Rehabil 1985;66:69–72.[Medline]
9. Kupperman HS, Blatt MHG, Wiesbader H, Filler W. Comparative clinical evaluation of estrogenic preparations by menopausal and amennorheal indices. J Clin Endocrinol 1953;13:688–703.
10. Ledin T, Kronhed A, Möller C, Möller M, Ödkvist L, Olsson B. Effects of balance training in elderly evaluated by clinical tests and dynamic posturography. J Vest Res 1991;1:129–38.
11. Ledin T, Gupta A, Ödkvist L, Lennmarken C. Dynamic and randomized perturbed posturography after nitrous oxide exposure. Proceedings of the XVth World Congress of Otorhinolaryngology, Istanbul, Turkey; Vol III. 1993:1217–20.
12. Hammar M, Ekblad S, Lönnberg B, Berg G, Lindgren R, Wyon Y. Postmenopausal women without previous or current vasomotor symptoms do not flush after abruptly abandoning estrogen replacement therapy. Maturitas 1999;31:117–22.[Medline]
13. Mc Ewen BS. Ovarian steroids have diverse effects on brain structures and function. In: Berg G, Hammar M, eds. The modern management of the menopause. London: Parthenon Publishing, 1994:269–78.
14. WHO Study group. Osteoporotic fractures. In: Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. Report of a WHO study group. WHO Technical Report Series 843; Geneva: Office of Publications World Health Organization, 1994:7–10.
15. Thomsen K, Gotfredsen A, Christiansen C. Is postmenopausal bone loss an age-related phenomenon? Calcif Tissue Int 1986;39: 123–7.[Medline]
16. Armstrong AL, Oborne J, Coupland CA, Macperson MB, Bassey EJ, Wallace WA. Effects of hormone replacement therapy on muscle performance in post-menopausal women. Clin Sci (Colch) 1996;91:685–90.[Medline]
17. Brown M, Birge SJ, Kohrt WM. Hormone replacement therapy does not augment gains in muscle strength or fat-free mass in response to weight-bearing exercise. J Gerontol A Biol Sci Med Sci 1997;52:166–70.
18. Naessen T, Lindmark B, Larsen H-C. Better postural balance in elderly women receiving estrogens. Am J Obstet Gynecol 1997;177: 412–6.[Medline]
19. Goebel JA, Birge SJ, Price SC, Hanson JM, Fishel DG. Estrogen replacement therapy and stability in the elderly. Am J Otolaryngol 1995;16:470–4.
20. Seeley DG, Cauley JA, Grady D, Browner WS, Nevitt MC, Cummings SR. Is postmenopausal estrogen therapy associated with neuromuscular function or falling in elderly women? Arch Intern Med 1995;155:293–9.[Abstract]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
