- Email: [email protected]
Effect of 10 years' hormone replacement therapy on bone mineral content in postmenopausal women
Bone Vol. 19, No. 5, Supplement November 1996:191S-193S
ELSEVIER
Effect of 10 Years' Hormone Replacement Therapy on Bone Mineral Content in Postmenopausal Women P. E I K E N , N. K O L T H O F F , and S. P O R S N I E L S E N Department (~f Clinical Physiology, Hillerod Hospital, Hillerod, Denmark
with a consequent increase in bone fragility and fracture risk. The site of osteoporotic fracture varies with age, with the incidence of Colles' fracture and vertebral fracture increasing soon after the menopause, probably as a result of the rapid loss of cancellous bone at this time. In contrast, the incidence of fracture of the hip, which contains large amounts of both cortical and cancellous bone, shows a more gradual increase, peaking at about the age of 80 years. Postmenopausal osteoporosis is a major cause of morbidity and mortality in women. Estimates suggest that at least 20% of hip fracture patients die within 1 year of injury, 5 with a further 50% requiring long-term c a r e Y Estrogen deficiency is considered to be the main factor leading to bone loss in postmenopausal women. A causal role of low circulating estrogen is suggested by the rapid and consistent loss of bone seen following both artificial and natural menopause, and by the protective effect of estrogen administration on bone mass in postmenopansal women as well as premenopausal women who have undergone ovariectomy, t6'~7 Bone mineral density is estimated to decrease by 2% each year during the first 5 years after the menopause, followed by an annual loss of approximately 1% for the rest of the woman's life. This translates into a reduction in overall bone mineral density of around 30% between the ages of 50 and 80. 6 Epidemiological evidence suggests a relationship between substantial vertebral bone loss and the incidence of vertebral crush fractures, ~8 as well as fractures of the hip. 15 It is well documented that hormone replacement therapy (HRT) initiated soon after the menopause reduces or reverses the loss of bone that normally occurs daring these years. 3'9A°'17 Moreover, numerous epidemiological and observational studies have shown a reduction in the incidence of osteoporotic fractures in postmenopansal women receiving HRT. 9'~e-~5,e°'2e'28 It is generally accepted that HRT should be started within the first 10 years of the menopause, when bone loss is greatest, to provide maximum prophylaxis against the risk of osteoporotic fractures in later life. At present, however, HRT is rarely sustained indefinitely in postmenopausal women, typically being given for periods from 9 months to 5 years. Data demonstrate that the beneficial effects of estrogen on bone are maintained for as long as therapy is continued; however, discontinuation of treatment results in an immediate resumption of bone loss at a rate comparable to that seen in untreated controls. 4 This suggests that HRT initiated at the time of the menopause should be sustained in the long term to afford optimum protection. Little is currently known about the long-term effects of HRT on bone mineral density. Long-term compliance with HRT tends to be poor, largely due to the monthly withdrawal bleeding seen with traditional sequential HRT regimens. 2"j~'24 This can be avoided by use of continuous combined HRT, with higher long-
The purpose of this study was to investigate the effects of 10 years of hormone replacement therapy (HRT) in postmenopausal women on bone mineral density of the lumbar spine (L-BMD) and bone mineral content of the distal forearm (F-BMC). A total of 151 women were enrolled in the study, 100 of whom were randomized to receive oral HRT (equally divided between a continuous combined and a sequential treatment regimen), with the remaining 51 receiving placebo or no treatment. The study was double-blind for the first 24 months, followed by 8 years of open-label follow-up. Total treatment duration was 10 years. At the end of 10 years, 38% of women randomized to continuous combined HRT remained on therapy compared with 22% of those who had received sequential HRT and 49% of the untreated group. A further 18% of women originally randomized to HRT had switched to other regimens. After 10 years of therapy, LBMD was found to be significantly higher in HRT-treated women than in those who remained untreated (14.5%; p < 0.001), corresponding to an increase in L-BMD of 13.1% from baseline values on HRT compared with a reduction in L-BMD of 4.7% without therapy. L-BMD increased by 15.9% in women receiving continuous combined therapy compared with 11.1% in those on sequential HRT; however, intergroup differences were not statistically significant. FBMC decreased by 0.7% over the 10 year period in the HRT treatment groups compared with a reduction of 17.6% in untreated women (p < 0.001). Mean F-BMC was 20.3% higher in women who had received HRT than in those who had not received therapy at the end of the 10 year follow-up. In conclusion, 10 years of treatment with HRT resulted in a substantial increase in L-BMD, with F-BMC also significantly higher in the HRT group than in untreated women. These results confirm that long-term HRT exerts a continuous effect against bone loss in postmenopausal women. (Bone
19:191S-193S; 1996) © 1996 by Elsevier Science Inc. Key Words: Menopause; Osteoporosis; Estrogen; Progestogen; Hormone replacement therapy; Bone mineral density.
Introduction Postmenopausal osteoporosis is a serious age-related disease that affects millions of women throughout the world. The condition is characterized by low bone mass and deterioration of bone tissue,
Address fi)r correspondence and reprints: Dr. Pia Eiken, Department of Clinical Physiology, Hillerod Hospital. 3400 HillerOd, Denmark. © 1996 by Elsevier Science Inc. All rights reserved.
191S
8756-3282/96/515.00 PII $8756-3282(96)00266-9
192S
P. Eiken et al. Long-term effects of ttRT
term compliance seen with such regimens compared with sequential HRT. 8 The purpose of this study was to evaluate the magnitude and duration of effect of 10 years of treatment with HRT versus no treatment on bone mineral density of the lumbar spine (L-BMD) and distal forearm in postmenopausal women. Results after 1 and 5 years of therapy have been published previously. 19"23 The preliminary analysis of the 10 year data is presented here.
Study Methodology
Bone Vol. 19, No. 5, Supplement November 1996:191S-193S L-BMD and F-BMC measured at 6 to 12 month intervals during the first 5 years of follow-up and then again after 8 and 10 years of therapy. After 10 years, L-BMD was measured using a dual X-ray absorptiometry bone densitometer (Hologic QDR-2000; Hologic Inc., Waltham, MA, USA). The long-term precision error of phantom studies was 0.35 %7 The scanner was calibrated to the initial scanner to avoid any significant differences in the mean bone mineral density values. Phantom analysis revealed no significant differences between mean values. F-BMC was measured using the same equipment as for previous assessments.
Patients and Study Design
Results
Ten years ago, 5800 women living in the hospital catchment area of HillerCd Hospital who were born between 1930 and 1933 were invited to enter a longitudinal study of vertebral and forearm bone loss and its possible prevention. 19'21'23 Only early postmenopausal women (defined as last vaginal bleeding more than 6 months and less than 24 months earlier) with no radiographic evidence of osteoarthritis or vertebral fractures were eligible to participate. Exclusion criteria included previous or current breast cancer, endometrial carcinoma, thromboembolic disease, parenchymatous liver disease, chronic pancreatitis, intestinal malabsorption and diabetes mellitus, obesity, and any other concurrent disease associated with abnormal bone turnover. Patients receiving treatment with estrogens, progestogens, corticosteroids, fluoride, vitamin D, or drugs known to provoke liver enzyme induction were also excluded from participation. After a 6 month run-in period, a total of 151 women were randomized to three groups by block randomization. All were given tablets of identical color and size in 28 day cycles. Group 1 (n = 50) received estrogen and progestogen continuously (one tablet dally containing 2 mg estradiol and 1 mg norethisterone acetate; Kliogest®, Novo Nordisk, Denmark). Group 2 (n = 50) took estrogen and progestogen sequentially (one tablet daily contalning 2 mg estradiol for the first 12 days of each 28 day cycle, with one tablet daily containing 2 mg estradiol and 1 mg norethisterone acetate for the next 10 days, followed by one tablet daily containing 1 mg estradiol for the remaining 6 days of the treatment cycle; Trisequens®, Novo Nordisk, Denmark). Group 3 (n = 51) received placebo tablets for the first 2 years, followed by no treatment. Treatment duration was 10 years. The study was double blind for the first 24 months, and after this time, all women were invited to continue on open-label medication. All women provided their informed consent to the study. The study was carried out in accordance with the Helsinki II Declaration and was approved by the local Ethical Committee and the Danish National Health Board.
Only women who completed l0 years of treatment were included in the efficacy analysis. After 10 years, 39 women were still on HRT (19 on the original continuous combined regimen, 11 on sequential therapy, and 9 on other HRT regimens), with 25 of the untreated group remaining on follow-up. After l0 years of treatment, mean L-BMD was found to be significantly higher in HRT-treated women than in those who remained untreated (p < 0.001) (Table 1), This corresponds to an increase in L-BMD of 13.1% from baseline values during treatment with HRT, with mean L-BMD in HRT-treated subjects being 14.5% higher than that of untreated women after 10 years of therapy (p < 0.001). Analysis by HRT treatment regimen revealed that meafl L-BMD increased by 15.9% in women who received continuous combined therapy over the 10 years of treatment compared with an increase of 11.l% in those who received sequential therapy. L-BMD decreased significantly in untreated women during the 10 year follow-up (4.7%; p < 0.01). Analysis of changes in bone density on therapy over time revealed a significant increase in L-BMD of 8.0% in women treated with continuous combined HRT between years 5 and l0 compared with a significant increase of 8.1% in those who received sequential therapy and 3.2% (n.s.) among untreated women during this time. F-BMC was seen to decrease in both the HRT and untreated groups over the 10 years, by 0.7% and 17.6%, respectively (p < 0.001) (Table 1), with no significant intergroup differences between HRT treatment regimens. After 10 years of therapy, mean F-BMC was 20.3% higher in women who had received HRT compared with those who received no therapy (p < 0.001). A mean increase in body weight of 4.3 kg was seen in patients who received HRT over the 10 year period of follow-up, compared with an increase of 3.2 kg in untreated controls. Intergroup differences in body weight were not statistically significant.
Assessments L-BMD was measured at baseline by planar anterior-posterior 153-Gd dual-photon absorptiometry of the second, third, and fourth lumbar vertebrae (L2-4, transverse processes excluded). The short-ternl precision error averaged 0.3% for bone powder standards and 1% for normal, nonobese subjects without osteoporosis.19 Forearm bone mineral content (F-BMC) was measured immediately after L-BMD at this time using a commercial I- 125 single-photon absorptiometer (Gammatec, V~erlCse, Denmark). Measurements began at a 10 mm distance between the radius and the ulna. Six scans proximal to this site were used for analysis. The short-term precision error for F-BMC was below 1.0% in nonobese women without osteoporosis. 23 Up to 12 measurements were made per woman during the 10 year study, with
Discussion The results of this study clearly demonstrate that HRT is effective for the long-term prevention of osteoporosis in postmenopausal women. At the end of 10 years, L-BMD was found to be significantly higher in HRT-treated women than in those who Table 1. Mean L-BMD (g/cm z) and mean F-BMC (g) at baseline and after 10 years of therapy L-BMD mean (SD)
HRT Untreated
F-BMC mean (SD)
Baseline
Year 10
Baseline
Year 10
0.962 (0.140) 0.998 (0.132)
1.089 (0.186) 0.951 (0.149)
1.191 (0.186) 1.194 (0.166)
1.183 (0.189) 0.984 (0.156)
Bone Vol. 19, No. 5, Supplement N o v e m b e r 1996:191S 193S
remained untreated (14.5%), corresponding to an increase in LBMD of 13.1% from baseline values on HRT compared with a reduction in L-BMD of 4.7% without therapy. In contrast to previous studies which have suggested that increases in bone mass tend to plateau after approximately 18 months of therapy, in this trial L-BMD increased continuously over the 10 years of follow-up in women receiving HRT, suggesting a continued anabolic effect, at least at the lumbar spine. Further analysis revealed that L-BMD increased by 15.9% in women receiving continuous combined therapy compared with 11.1% in those on sequential HRT; however, intergroup differences did not attain statistical significance. Estrogen replacement therapy is recognized as the most effective means of preventing loss of bone mass in postmenopausal women. 27 However, monotherapy with norethisterone acetate may also have a beneficial effect on bone. 1 This suggests that the differences in L-BMD seen between the two HRT regimens in the present study are most likely due to the greater dose of norethisterone received during continuous combined therapy. F-BMC decreased by 0.7% over the 10 year period in the HRT treatment groups, compared with a reduction of 17.6% in untreated women, with no significant differences in F-BMC between the two HRT treatment regimens during the study. Mean F-BMC was 20.3% higher in women who had received HRT than in those who had not received therapy at the end of the 10 year follow-up. Although the pattern of the skeletal response to HRT is not known in detail, it is widely accepted that HRT has a more pronounced effect on the axial than the peripheral skeleton. 1o,26 This is supported by the results of the present study and is most likely due to the relatively high turnover in the vertebrae, which consist mainly of cancellous bone, compared with the predominantly cortical forearm. A commonly used theoretical fracture risk limit for the lumbar spine is a mean L-BMD below 0.9 g/cm=. Applied on the actual values seen in this trial, this translates into a reduction in fracture risk from 33% at baseline to 13% at the end of therapy ( - 6 1 % ) in women who received HRT, compared with an increase from 28% to 40% (+43%) in those who remained untreated. The results would therefore appear to confirm that longterm treatment with HRT can protect postmenopausal women against osteoporotic fractures. Conclusion In summary, l0 years of treatment with HRT resulted in a substantial increase in L-BMD, with preservation of F-BMC. The available data suggest that increases in L-BMD may be greater in women receiving continuous combined therapy than those on a sequential regimen; however, intergroup differences were not statistically significant. Long-term compliance was also higher with the continuous regimen than with sequential therapy, presumably due to the absence of a monthly bleed. These findings confirm that continuous combined HRT is well suited for the long-term prevention of osteoporosis in postmenopausal women. This approach may also provide the potential for hormonal intervention in older women at risk of this disease. References 1. Abdalla, H. I., Hart, B. M., Lindsay, R., et al. Prevention of bone mineral loss in postmenopausal women with norelhisterone. Obstet Gynecol 66:789-792; 1985. 2. Barentsen, R., Groeneveld, F. P. M. J., Bareman, F. P., et al. Women's opinion
P. Eiken et al. Long-terrn effects of H R T
3.
4.
5. 6. 7. 8. 9. 10.
11. 12.
13.
14. 15.
16.
17. 18. 19.
20.
21.
22. 23.
24. 25. 26.
27.
28.
193S
on withdrawal bleeding with hormone replacement therapy. Eur J Obstet Gynecol Reprod Biol 51:203-207; 1993. Christiansen, C., Christiansen, M. S., McNair, P. L., et al. Prevention of early postmenopausal bone loss: Conducted 2-year study of 315 normal females. Eur J Clin Invest 10:273-279; 1980. Christianscn, C., Christiansen, M. S., and Transb¢l, I. B. Bone mass in postmenopausal women after withdrawal of oestrogen/gestagen replacement therapy. Lancet 1:459461; 1981. Consensus Development Conference. Prophylaxis and treatment of osteoporosis. Am J Med 90:107 110; 1991. Dennison, E. and Cooper, C. The epidemiology of osteoporosis. Br J Clin Pract 50:33-36; 1996. Eiken, P., B~,renholdt, O., Bj0m Jensen, L., et al. Switching from DXA pencil beam to fan-beam. 1: Studies in vitro at four centers. Bone 15:667-670: 1994. Eiken, P. and Kolthoff, N. Compliance with long-term oral hormonal replacement therapy. Matmitas 22:97-103; 1995. Ettinger, B., Genant, H. K., and Cann, C. E. Long-term estrogen replacement therapy prevents bone loss and fractures. Ann Intern Med 102:31~324; 1985. Gotfredsen, A., Nilas, L , Riis, B. J., et at. Bone changes occurring spontaneously and caused by estrogen in early postmenopansal women: A local or generalised phenomenon? Br Med J 292:1098-2000; 1986. Hahn, R.G. Compliance considerations with estrogen replacement: Withdrawal bleeding and other factors. Am J Obstet Gyuecol 161 : 1854-1858; 1989. Hutchinson, T. A., Polansky, S. M., and Feinstein, A. R. Post-menopausal oestrogen protects against fractures of the hip and distal radius. A case-control study. Lancet 2:705 709; 1979. Johnson, R. E. and Sprecht, E. E. The risk of hip fracture in postmenopausal women with or without estrogen drug exposure. Am J Public Health 71:138144; 1981. Keil, D. P., Felson, D. T., Anderson, J.J., et al. Hip fracture and the use of estrogens in postmenopausal women. N Engl J Med 317:1169 1174; 1987. Kreiger, N., Kelsey, J. L., Holford, T. R., and O'Connor, T. An epidemiologic study of hip fracture in postmenopausal women. Am J Epidemiol 116:141 148; 1982. Lindsay, R., Aitken, J. M., Anderson, J. B., et al. Long-term prevention of postmenopausal osteoporosis by oestrogen treatment. Lancet 1:1038-1041; 1976. Lindsay, R., Hart, D. M., Forrest, C., and Baird, C. Prevention of spinal osteoporosis in oophorectomised women. Lancet 2:1151 1153; 1980. Mazess, R. B. Bone density in diagnosis of osteoporosis: Thresholds and breakpoints. Calcif Tissue Int 41:117 118; 1987. Munk-Jensen, N., Pors Nielsen, S., Obel, E. B., and Bonne Eriksen, P. Reversal of postmenopausal vertebral bone loss by oeslrogen and progestogen: A double blind placebo controlled study. Br Med J 296:1150-1152; 1988. Naessrn, T., Persson, l., Adami, H.-O., et al. Hormone replacement therapy and risk of first hip fracture. A prospective, population-based cohort study. Ann Intern Med 113:95-103; 1990. Obel, E. B., Munk-Jensen, N., Svenstrup, B., et al. A two-year double-blind controlled study of the clinical effect of combined and sequential postmenopausal replacement therapy and steroid metabolism during treatment. Maturitas 16:13 21; 1993. Paganini-Hill, A., Ross. R. K., Gerkins, V. R., et al. Menopausal estrogen therapy and hip fractures. Ann Intern Med 95:28-31; 198l. Pots Nielsen, S., Bfirenholdt, O., Hermansen, F., and Munk-Jensen, N. Magnitude and pattern of skeletal response to long term continuous and cyclic sequential oestrogerdprogestin treatment. Br J Obstet Gyuaecol 10l:319 324; 1994. Rees, M. On menstrual bleeding with hormone replacement therapy. Lancet 343:250; 1994. Rowe, R. Preventative strategies: ls current clinical practice effective for bones? Br J Clin Pract 50:47~-9; 1996. Riis, B. J., Thomsen. K., Str0m. V., and Christiansen, C. The eflect of percutaneous estradiol and natural progesterone on postmenopausal bone loss. Am J Obstet Gynecol 156:6[-65; 1987. Rozenbaum, H. and Birkh~user, M. H. Consensus Propositions. Birkh~user M. H. and Rozenbaum H., Eds. Menopause, European Consensus Development Conference, Montreux, Switzerland. Editions ESKA, Paris; pp. 293-303, 1996. Weiss, N. S., Ure, C. L., Ballard, J. H., et at. Decreased risk of fractures of the hip and lower forearm with postmenopausal use of estrogen. N Engl J Med 303:1195 1198; 1980.
ELSEVIER
Effect of 10 Years' Hormone Replacement Therapy on Bone Mineral Content in Postmenopausal Women P. E I K E N , N. K O L T H O F F , and S. P O R S N I E L S E N Department (~f Clinical Physiology, Hillerod Hospital, Hillerod, Denmark
with a consequent increase in bone fragility and fracture risk. The site of osteoporotic fracture varies with age, with the incidence of Colles' fracture and vertebral fracture increasing soon after the menopause, probably as a result of the rapid loss of cancellous bone at this time. In contrast, the incidence of fracture of the hip, which contains large amounts of both cortical and cancellous bone, shows a more gradual increase, peaking at about the age of 80 years. Postmenopausal osteoporosis is a major cause of morbidity and mortality in women. Estimates suggest that at least 20% of hip fracture patients die within 1 year of injury, 5 with a further 50% requiring long-term c a r e Y Estrogen deficiency is considered to be the main factor leading to bone loss in postmenopausal women. A causal role of low circulating estrogen is suggested by the rapid and consistent loss of bone seen following both artificial and natural menopause, and by the protective effect of estrogen administration on bone mass in postmenopansal women as well as premenopausal women who have undergone ovariectomy, t6'~7 Bone mineral density is estimated to decrease by 2% each year during the first 5 years after the menopause, followed by an annual loss of approximately 1% for the rest of the woman's life. This translates into a reduction in overall bone mineral density of around 30% between the ages of 50 and 80. 6 Epidemiological evidence suggests a relationship between substantial vertebral bone loss and the incidence of vertebral crush fractures, ~8 as well as fractures of the hip. 15 It is well documented that hormone replacement therapy (HRT) initiated soon after the menopause reduces or reverses the loss of bone that normally occurs daring these years. 3'9A°'17 Moreover, numerous epidemiological and observational studies have shown a reduction in the incidence of osteoporotic fractures in postmenopansal women receiving HRT. 9'~e-~5,e°'2e'28 It is generally accepted that HRT should be started within the first 10 years of the menopause, when bone loss is greatest, to provide maximum prophylaxis against the risk of osteoporotic fractures in later life. At present, however, HRT is rarely sustained indefinitely in postmenopausal women, typically being given for periods from 9 months to 5 years. Data demonstrate that the beneficial effects of estrogen on bone are maintained for as long as therapy is continued; however, discontinuation of treatment results in an immediate resumption of bone loss at a rate comparable to that seen in untreated controls. 4 This suggests that HRT initiated at the time of the menopause should be sustained in the long term to afford optimum protection. Little is currently known about the long-term effects of HRT on bone mineral density. Long-term compliance with HRT tends to be poor, largely due to the monthly withdrawal bleeding seen with traditional sequential HRT regimens. 2"j~'24 This can be avoided by use of continuous combined HRT, with higher long-
The purpose of this study was to investigate the effects of 10 years of hormone replacement therapy (HRT) in postmenopausal women on bone mineral density of the lumbar spine (L-BMD) and bone mineral content of the distal forearm (F-BMC). A total of 151 women were enrolled in the study, 100 of whom were randomized to receive oral HRT (equally divided between a continuous combined and a sequential treatment regimen), with the remaining 51 receiving placebo or no treatment. The study was double-blind for the first 24 months, followed by 8 years of open-label follow-up. Total treatment duration was 10 years. At the end of 10 years, 38% of women randomized to continuous combined HRT remained on therapy compared with 22% of those who had received sequential HRT and 49% of the untreated group. A further 18% of women originally randomized to HRT had switched to other regimens. After 10 years of therapy, LBMD was found to be significantly higher in HRT-treated women than in those who remained untreated (14.5%; p < 0.001), corresponding to an increase in L-BMD of 13.1% from baseline values on HRT compared with a reduction in L-BMD of 4.7% without therapy. L-BMD increased by 15.9% in women receiving continuous combined therapy compared with 11.1% in those on sequential HRT; however, intergroup differences were not statistically significant. FBMC decreased by 0.7% over the 10 year period in the HRT treatment groups compared with a reduction of 17.6% in untreated women (p < 0.001). Mean F-BMC was 20.3% higher in women who had received HRT than in those who had not received therapy at the end of the 10 year follow-up. In conclusion, 10 years of treatment with HRT resulted in a substantial increase in L-BMD, with F-BMC also significantly higher in the HRT group than in untreated women. These results confirm that long-term HRT exerts a continuous effect against bone loss in postmenopausal women. (Bone
19:191S-193S; 1996) © 1996 by Elsevier Science Inc. Key Words: Menopause; Osteoporosis; Estrogen; Progestogen; Hormone replacement therapy; Bone mineral density.
Introduction Postmenopausal osteoporosis is a serious age-related disease that affects millions of women throughout the world. The condition is characterized by low bone mass and deterioration of bone tissue,
Address fi)r correspondence and reprints: Dr. Pia Eiken, Department of Clinical Physiology, Hillerod Hospital. 3400 HillerOd, Denmark. © 1996 by Elsevier Science Inc. All rights reserved.
191S
8756-3282/96/515.00 PII $8756-3282(96)00266-9
192S
P. Eiken et al. Long-term effects of ttRT
term compliance seen with such regimens compared with sequential HRT. 8 The purpose of this study was to evaluate the magnitude and duration of effect of 10 years of treatment with HRT versus no treatment on bone mineral density of the lumbar spine (L-BMD) and distal forearm in postmenopausal women. Results after 1 and 5 years of therapy have been published previously. 19"23 The preliminary analysis of the 10 year data is presented here.
Study Methodology
Bone Vol. 19, No. 5, Supplement November 1996:191S-193S L-BMD and F-BMC measured at 6 to 12 month intervals during the first 5 years of follow-up and then again after 8 and 10 years of therapy. After 10 years, L-BMD was measured using a dual X-ray absorptiometry bone densitometer (Hologic QDR-2000; Hologic Inc., Waltham, MA, USA). The long-term precision error of phantom studies was 0.35 %7 The scanner was calibrated to the initial scanner to avoid any significant differences in the mean bone mineral density values. Phantom analysis revealed no significant differences between mean values. F-BMC was measured using the same equipment as for previous assessments.
Patients and Study Design
Results
Ten years ago, 5800 women living in the hospital catchment area of HillerCd Hospital who were born between 1930 and 1933 were invited to enter a longitudinal study of vertebral and forearm bone loss and its possible prevention. 19'21'23 Only early postmenopausal women (defined as last vaginal bleeding more than 6 months and less than 24 months earlier) with no radiographic evidence of osteoarthritis or vertebral fractures were eligible to participate. Exclusion criteria included previous or current breast cancer, endometrial carcinoma, thromboembolic disease, parenchymatous liver disease, chronic pancreatitis, intestinal malabsorption and diabetes mellitus, obesity, and any other concurrent disease associated with abnormal bone turnover. Patients receiving treatment with estrogens, progestogens, corticosteroids, fluoride, vitamin D, or drugs known to provoke liver enzyme induction were also excluded from participation. After a 6 month run-in period, a total of 151 women were randomized to three groups by block randomization. All were given tablets of identical color and size in 28 day cycles. Group 1 (n = 50) received estrogen and progestogen continuously (one tablet dally containing 2 mg estradiol and 1 mg norethisterone acetate; Kliogest®, Novo Nordisk, Denmark). Group 2 (n = 50) took estrogen and progestogen sequentially (one tablet daily contalning 2 mg estradiol for the first 12 days of each 28 day cycle, with one tablet daily containing 2 mg estradiol and 1 mg norethisterone acetate for the next 10 days, followed by one tablet daily containing 1 mg estradiol for the remaining 6 days of the treatment cycle; Trisequens®, Novo Nordisk, Denmark). Group 3 (n = 51) received placebo tablets for the first 2 years, followed by no treatment. Treatment duration was 10 years. The study was double blind for the first 24 months, and after this time, all women were invited to continue on open-label medication. All women provided their informed consent to the study. The study was carried out in accordance with the Helsinki II Declaration and was approved by the local Ethical Committee and the Danish National Health Board.
Only women who completed l0 years of treatment were included in the efficacy analysis. After 10 years, 39 women were still on HRT (19 on the original continuous combined regimen, 11 on sequential therapy, and 9 on other HRT regimens), with 25 of the untreated group remaining on follow-up. After l0 years of treatment, mean L-BMD was found to be significantly higher in HRT-treated women than in those who remained untreated (p < 0.001) (Table 1), This corresponds to an increase in L-BMD of 13.1% from baseline values during treatment with HRT, with mean L-BMD in HRT-treated subjects being 14.5% higher than that of untreated women after 10 years of therapy (p < 0.001). Analysis by HRT treatment regimen revealed that meafl L-BMD increased by 15.9% in women who received continuous combined therapy over the 10 years of treatment compared with an increase of 11.l% in those who received sequential therapy. L-BMD decreased significantly in untreated women during the 10 year follow-up (4.7%; p < 0.01). Analysis of changes in bone density on therapy over time revealed a significant increase in L-BMD of 8.0% in women treated with continuous combined HRT between years 5 and l0 compared with a significant increase of 8.1% in those who received sequential therapy and 3.2% (n.s.) among untreated women during this time. F-BMC was seen to decrease in both the HRT and untreated groups over the 10 years, by 0.7% and 17.6%, respectively (p < 0.001) (Table 1), with no significant intergroup differences between HRT treatment regimens. After 10 years of therapy, mean F-BMC was 20.3% higher in women who had received HRT compared with those who received no therapy (p < 0.001). A mean increase in body weight of 4.3 kg was seen in patients who received HRT over the 10 year period of follow-up, compared with an increase of 3.2 kg in untreated controls. Intergroup differences in body weight were not statistically significant.
Assessments L-BMD was measured at baseline by planar anterior-posterior 153-Gd dual-photon absorptiometry of the second, third, and fourth lumbar vertebrae (L2-4, transverse processes excluded). The short-ternl precision error averaged 0.3% for bone powder standards and 1% for normal, nonobese subjects without osteoporosis.19 Forearm bone mineral content (F-BMC) was measured immediately after L-BMD at this time using a commercial I- 125 single-photon absorptiometer (Gammatec, V~erlCse, Denmark). Measurements began at a 10 mm distance between the radius and the ulna. Six scans proximal to this site were used for analysis. The short-term precision error for F-BMC was below 1.0% in nonobese women without osteoporosis. 23 Up to 12 measurements were made per woman during the 10 year study, with
Discussion The results of this study clearly demonstrate that HRT is effective for the long-term prevention of osteoporosis in postmenopausal women. At the end of 10 years, L-BMD was found to be significantly higher in HRT-treated women than in those who Table 1. Mean L-BMD (g/cm z) and mean F-BMC (g) at baseline and after 10 years of therapy L-BMD mean (SD)
HRT Untreated
F-BMC mean (SD)
Baseline
Year 10
Baseline
Year 10
0.962 (0.140) 0.998 (0.132)
1.089 (0.186) 0.951 (0.149)
1.191 (0.186) 1.194 (0.166)
1.183 (0.189) 0.984 (0.156)
Bone Vol. 19, No. 5, Supplement N o v e m b e r 1996:191S 193S
remained untreated (14.5%), corresponding to an increase in LBMD of 13.1% from baseline values on HRT compared with a reduction in L-BMD of 4.7% without therapy. In contrast to previous studies which have suggested that increases in bone mass tend to plateau after approximately 18 months of therapy, in this trial L-BMD increased continuously over the 10 years of follow-up in women receiving HRT, suggesting a continued anabolic effect, at least at the lumbar spine. Further analysis revealed that L-BMD increased by 15.9% in women receiving continuous combined therapy compared with 11.1% in those on sequential HRT; however, intergroup differences did not attain statistical significance. Estrogen replacement therapy is recognized as the most effective means of preventing loss of bone mass in postmenopausal women. 27 However, monotherapy with norethisterone acetate may also have a beneficial effect on bone. 1 This suggests that the differences in L-BMD seen between the two HRT regimens in the present study are most likely due to the greater dose of norethisterone received during continuous combined therapy. F-BMC decreased by 0.7% over the 10 year period in the HRT treatment groups, compared with a reduction of 17.6% in untreated women, with no significant differences in F-BMC between the two HRT treatment regimens during the study. Mean F-BMC was 20.3% higher in women who had received HRT than in those who had not received therapy at the end of the 10 year follow-up. Although the pattern of the skeletal response to HRT is not known in detail, it is widely accepted that HRT has a more pronounced effect on the axial than the peripheral skeleton. 1o,26 This is supported by the results of the present study and is most likely due to the relatively high turnover in the vertebrae, which consist mainly of cancellous bone, compared with the predominantly cortical forearm. A commonly used theoretical fracture risk limit for the lumbar spine is a mean L-BMD below 0.9 g/cm=. Applied on the actual values seen in this trial, this translates into a reduction in fracture risk from 33% at baseline to 13% at the end of therapy ( - 6 1 % ) in women who received HRT, compared with an increase from 28% to 40% (+43%) in those who remained untreated. The results would therefore appear to confirm that longterm treatment with HRT can protect postmenopausal women against osteoporotic fractures. Conclusion In summary, l0 years of treatment with HRT resulted in a substantial increase in L-BMD, with preservation of F-BMC. The available data suggest that increases in L-BMD may be greater in women receiving continuous combined therapy than those on a sequential regimen; however, intergroup differences were not statistically significant. Long-term compliance was also higher with the continuous regimen than with sequential therapy, presumably due to the absence of a monthly bleed. These findings confirm that continuous combined HRT is well suited for the long-term prevention of osteoporosis in postmenopausal women. This approach may also provide the potential for hormonal intervention in older women at risk of this disease. References 1. Abdalla, H. I., Hart, B. M., Lindsay, R., et al. Prevention of bone mineral loss in postmenopausal women with norelhisterone. Obstet Gynecol 66:789-792; 1985. 2. Barentsen, R., Groeneveld, F. P. M. J., Bareman, F. P., et al. Women's opinion
P. Eiken et al. Long-terrn effects of H R T
3.
4.
5. 6. 7. 8. 9. 10.
11. 12.
13.
14. 15.
16.
17. 18. 19.
20.
21.
22. 23.
24. 25. 26.
27.
28.
193S
on withdrawal bleeding with hormone replacement therapy. Eur J Obstet Gynecol Reprod Biol 51:203-207; 1993. Christiansen, C., Christiansen, M. S., McNair, P. L., et al. Prevention of early postmenopausal bone loss: Conducted 2-year study of 315 normal females. Eur J Clin Invest 10:273-279; 1980. Christianscn, C., Christiansen, M. S., and Transb¢l, I. B. Bone mass in postmenopausal women after withdrawal of oestrogen/gestagen replacement therapy. Lancet 1:459461; 1981. Consensus Development Conference. Prophylaxis and treatment of osteoporosis. Am J Med 90:107 110; 1991. Dennison, E. and Cooper, C. The epidemiology of osteoporosis. Br J Clin Pract 50:33-36; 1996. Eiken, P., B~,renholdt, O., Bj0m Jensen, L., et al. Switching from DXA pencil beam to fan-beam. 1: Studies in vitro at four centers. Bone 15:667-670: 1994. Eiken, P. and Kolthoff, N. Compliance with long-term oral hormonal replacement therapy. Matmitas 22:97-103; 1995. Ettinger, B., Genant, H. K., and Cann, C. E. Long-term estrogen replacement therapy prevents bone loss and fractures. Ann Intern Med 102:31~324; 1985. Gotfredsen, A., Nilas, L , Riis, B. J., et at. Bone changes occurring spontaneously and caused by estrogen in early postmenopansal women: A local or generalised phenomenon? Br Med J 292:1098-2000; 1986. Hahn, R.G. Compliance considerations with estrogen replacement: Withdrawal bleeding and other factors. Am J Obstet Gyuecol 161 : 1854-1858; 1989. Hutchinson, T. A., Polansky, S. M., and Feinstein, A. R. Post-menopausal oestrogen protects against fractures of the hip and distal radius. A case-control study. Lancet 2:705 709; 1979. Johnson, R. E. and Sprecht, E. E. The risk of hip fracture in postmenopausal women with or without estrogen drug exposure. Am J Public Health 71:138144; 1981. Keil, D. P., Felson, D. T., Anderson, J.J., et al. Hip fracture and the use of estrogens in postmenopausal women. N Engl J Med 317:1169 1174; 1987. Kreiger, N., Kelsey, J. L., Holford, T. R., and O'Connor, T. An epidemiologic study of hip fracture in postmenopausal women. Am J Epidemiol 116:141 148; 1982. Lindsay, R., Aitken, J. M., Anderson, J. B., et al. Long-term prevention of postmenopausal osteoporosis by oestrogen treatment. Lancet 1:1038-1041; 1976. Lindsay, R., Hart, D. M., Forrest, C., and Baird, C. Prevention of spinal osteoporosis in oophorectomised women. Lancet 2:1151 1153; 1980. Mazess, R. B. Bone density in diagnosis of osteoporosis: Thresholds and breakpoints. Calcif Tissue Int 41:117 118; 1987. Munk-Jensen, N., Pors Nielsen, S., Obel, E. B., and Bonne Eriksen, P. Reversal of postmenopausal vertebral bone loss by oeslrogen and progestogen: A double blind placebo controlled study. Br Med J 296:1150-1152; 1988. Naessrn, T., Persson, l., Adami, H.-O., et al. Hormone replacement therapy and risk of first hip fracture. A prospective, population-based cohort study. Ann Intern Med 113:95-103; 1990. Obel, E. B., Munk-Jensen, N., Svenstrup, B., et al. A two-year double-blind controlled study of the clinical effect of combined and sequential postmenopausal replacement therapy and steroid metabolism during treatment. Maturitas 16:13 21; 1993. Paganini-Hill, A., Ross. R. K., Gerkins, V. R., et al. Menopausal estrogen therapy and hip fractures. Ann Intern Med 95:28-31; 198l. Pots Nielsen, S., Bfirenholdt, O., Hermansen, F., and Munk-Jensen, N. Magnitude and pattern of skeletal response to long term continuous and cyclic sequential oestrogerdprogestin treatment. Br J Obstet Gyuaecol 10l:319 324; 1994. Rees, M. On menstrual bleeding with hormone replacement therapy. Lancet 343:250; 1994. Rowe, R. Preventative strategies: ls current clinical practice effective for bones? Br J Clin Pract 50:47~-9; 1996. Riis, B. J., Thomsen. K., Str0m. V., and Christiansen, C. The eflect of percutaneous estradiol and natural progesterone on postmenopausal bone loss. Am J Obstet Gynecol 156:6[-65; 1987. Rozenbaum, H. and Birkh~user, M. H. Consensus Propositions. Birkh~user M. H. and Rozenbaum H., Eds. Menopause, European Consensus Development Conference, Montreux, Switzerland. Editions ESKA, Paris; pp. 293-303, 1996. Weiss, N. S., Ure, C. L., Ballard, J. H., et at. Decreased risk of fractures of the hip and lower forearm with postmenopausal use of estrogen. N Engl J Med 303:1195 1198; 1980.