MENOPAUSE IN women produces estrogen deficiency. Does Hormone-Replacement Therapy Prevent Fractures in Early Postmenopausal Women?

Transcription

1 JOURNAL OF BONE AND MINERAL RESEARCH Volume 17, Number 3, American Society for Bone and Mineral Research Does Hormone-Replacement Therapy Prevent Fractures in Early Postmenopausal Women? KAISA M. RANDELL, 1,2 RISTO J. HONKANEN, 2 HEIKKI KRÖGER, 3 and SEPPO SAARIKOSKI 1 ABSTRACT The purpose of this population-based prospective cohort study was to examine the effect of hormonereplacement therapy (HRT) on the risk of. The study population consisted of 7217 postmenopausal women aged years (mean, 53.3 years) at baseline from data taken from the Kuopio Osteoporosis Risk Factor and Prevention Study (OSTPRE) in Finland. We compared fracture incidences between HRT users and nonusers. A total of 679 (9.4%) women recorded validated during the 5-year follow-up. Of these, 268 (39%) women had a distal forearm fracture. Two thousand six hundred seventy women (37%) had used HRT >6 months during the follow-up one-half of them continuously. The relative risk, estimated as hazard ratio with Cox regression, was 0.69 (95% CI, ) for any fracture and 0.49 ( ) for distal forearm fracture among HRT users as compared with never-users. After adjusting for age, body mass index (BMI), number of chronic health disorders, fracture history, and time since menopause (independent risk factors) the corresponding risks were 0.67 ( ) and 0.53 ( ), respectively. The respective adjusted risks for continuous HRT users were 0.62 ( ) and 0.41 ( ). The adjusted risk of other than distal forearm fracture was 0.74 ( ). The results suggest that HRT has a beneficial effect on prevention of in general and on that of distal forearm fracture in particular in early postmenopausal women. (J Bone Miner Res 2002;17: ) Key words:, hormone-replacement therapy, menopause, women INTRODUCTION MENOPAUSE IN women produces estrogen deficiency leading to the increased risk of osteoporosis and osteoporotic. Hormone-replacement therapy (HRT) has proved to be effective for prevention of postmenopausal bone loss. (1 4) Previous studies about HRT and nonvertebral are few and have mostly been performed on elderly women (5,6) or have examined hip fracture risk related to HRT. (7 10) The effect of HRT on in younger women is less established. (11,12) Forearm fracture is the most common fracture in The authors have no conflict of interest. early postmenopausal women. However, studies about the HRT effect on forearm fracture are few and small. (12 15) In addition, a Swedish case-control study (16) and our prospective cohort study (17) on risk factors for distal forearm fracture suggest that HRT prevents this fracture. The purpose of this study was to evaluate the effect of HRT on clinically diagnosed bone in early postmenopausal women. MATERIALS AND METHODS Study population This study was part of a 5-year prospective populationbased cohort study: the Osteoporosis Risk Factor and Pre- 1 Department of Obstetrics and Gynecology, Kuopio University Hospital, Kuopio, Finland. 2 Research Institute of Public Health, University of Kuopio, Kuopio, Finland. 3 Department of Surgery, Kuopio University Hospital, Kuopio, Finland. 528

2 DOES HRT PREVENT FRACTURES? 529 TABLE 1. CHARACTERISTICS OF THE 7217 POSTMENOPAUSAL WOMEN ACCORDING TO HRT USE AT BASELINE IN 1989 During follow-up Characteristic Never (n 3335) Past only (n 1212) Part-time (n 1335) Continuous (n 1335) Total (n 7217) Means Age (mean [SD]/years) 54.0 (2.5) 53.3 (2.7)* 52.4 (2.8)* 52.5 (2.6)* 53.3 (2.7) Time since menopause (years) 5.16 (3.89) 3.91 (4.12)* 2.62 (3.64)* 2.83 (4.08)* 4.05 (4.07) Height (cm) (5.4) (5.3) (5.2)* (5.1)* (5.3) Weight (kg) 69.4 (12.5) 69.5 (12.3) 66.6 (10.1)* 65.6 (9.6)* 68.2 (11.7) BMI (kg/m 2 ) 27.0 (4.1) 27.0 (4.5) 25.5 (3.6)* 25.1 (3.4)* 26.3 (4.3) Dairy Ca intake (mg/day) 857 (413) 811 (384)* 811 (383)* 795 (360)* 829 (394) Proportions Menopause status 5 years ago (%) years ago (%) Within 2 years (%) Nulliparous (%) Bilateral oophorectomy (%; ad 1989) Hysterectomy (%; ad 1989) Smoking (%) Physically active (%) of health disorder (2 ; %) History of fracture (%) Use of calcium supplements (%) Comparison of each HRT group to never-used group: * p 0.05, analysis of variance; p 0.05, 2 -test. vention Study (OSTPRE) baseline postal inquiry was sent to all the women aged years who reside in Kuopio Province Eastern Finland in May Thirteen thousand one hundred women (92%) responded to the baseline inquiry. The 5-year follow-up inquiry was carried out in May Eleven thousand seven hundred ninety-eight women responded to both inquiries. A woman was regarded as postmenopausal if 6 months had elapsed since her last natural menstruation. Women whose menopause could not be defined because of the hysterectomy performed before menopause were excluded (n 1044). The final study population base of this study consisted of 7217 women who responded to both questionnaires and were postmenopausal in Variables The mailed baseline questionnaire included questions about fracture history, weight, height, health status, medications including calcium supplements, operations, HRT use, reproductive history, parity, gynecological operations, behavioral and medical risk factors, physical activity, smoking, and consumption of milk products. In the follow-up inquiry, HRT use and during the follow-up period were examined. Body mass index (BMI) was calculated as the ratio of weight in kilograms to height in square meters. Number of chronic health disorders (2 ) meant diseases diagnosed by a doctor. It was used as a dichotomous variable with a cut-off point of none or one versus two or more health disorders. A woman was regarded as physically active if she reported regular physical activity during leisure time. The dairy calcium (mg/day) intake was the sum of intakes from liquid milk products (120 mg/dl) and cheese (87 mg/slice). Fracture history meant sustained since the age of 15 years by the baseline questionnaire. Follow-up The participants were asked if they had sustained any bone diagnosed by a doctor during follow-up. They were requested to list all, with bone site, trauma mechanism, time of occurrence, and place of medical care. Reported follow-up were validated by medical records and radiographic reports. Only 18 (2.7%) were caused by car crash. These were not excluded. HRT In the baseline inquiry, the lifetime use (and indication) of HRT in years was recorded. The follow-up inquiry recorded the number of months in each year of follow-up that HRT had been used. Missing or nonlogical information was completed by telephone interview. Women were pooled into four groups according to HRT use: (1) never used, (2) past use (before baseline only), (3) part-time (more than 6 months) use during follow-up, and (4) continuous (over 4.5 years) use throughout the 5-year follow-up. Statistical analysis The statistical analyses were performed with the Statistical Package for Social Sciences (SPSS, Inc. Tokyo, Japan)

3 530 RANDELL ET AL. program. The differences in baseline characteristics between each HRT group and never-used HRT group were tested by using analysis of variance (ANOVA) to compare means of continuous variables and the 2 test to compare categorical variable distributions. The Cox proportional hazards regression model was used to assess the relation between HRT use and fracture risk. The relative risks are expressed as hazard ratios. Women who never had used HRT formed the reference group for all analyses. Time from June 1, 1989 to the first fracture was used as the dependent variable. RESULTS Baseline characteristics Baseline characteristics of the study subjects according to HRT use are shown in Table 1. The mean age of the study population was 53.3 years at baseline. Women who had used HRT, either part-time or continuously throughout the follow-up, were significantly younger, taller, and thinner compared with never-users. HRT users also were healthier, physically more active, and they had experienced less time since menopause. They also smoked less and had lower calcium intake but used more calcium supplements. The indications of HRT were distributed evenly among women who had or did not have bone. Of all the HRT users, 85% used it for menopausal symptoms, 8 12% for replacement therapy after bilateral oophorectomy, and the rest for other reasons including prevention of bone loss. Fracture incidence A total of 679 (9.4%) women had sustained a validated follow-up fracture. Of these, 268 (39%) women had a distal forearm fracture. The next most frequent fracture sites were ankle (n 108), rib (n 74), toe (n 32), and humerus (n 28). Because some women had sustained more than one fracture (in different sites), altogether, 782 validated were recorded. The numbers of are shown in the Table 2. Women who were 2 years past menopause at baseline sustained more than those who were within 2 years of menopause at baseline (10% vs. 8%; p 0.035). Distribution of according to trauma was as follows: fall on same level, 68.2%; fall from one level to another, 4.1%; car crash, 2.7%; bicycle accident, 6.2%; and other trauma, 18.8%. Impact of HRT on The risk of fracture was significantly lower in part-time or continuous users of HRT compared with never-users of HRT (Tables 3 and 4). The estimated relative risk for any fracture was 0.69 (95% CI, ) and 0.49 (95% CI, ) for distal forearm fracture among all HRT users as compared with those not using HRT. After adjusting for age, BMI, number of chronic health disorders, previous fracture, and time since menopause, which were independent risk factors for TABLE 2. FRACTURES AMONG 7217 POSTMENOPAUSAL WOMEN Site of fracture of Distal forearm/wrist 280 Ankle 108 Rib 74 Toe 32 Proximal humerus/shoulder 28 Elbow 27 Metatarsus 27 Finger 25 Metacarpus 20 Lumbar vertebral column 23 Shaft of fibula 15 Hip/proximal femur 13 Shaft of tibia/tibia fibula 12 Carpal bones 12 Head/face/skull 10 Thoracic vertebral column 9 Proximal tibia 9 Patella 8 Shoulder blade 8 Clavicle 7 Distal femur 7 Sacrum/coccyx 7 Shaft of radius/ulna 5 Shaft of humerus 4 Shaft of femur 4 Sternum 3 Cervical vertebral column 2 Tarsal bones 2 Pelvis 1 Total 782 any or distal forearm fracture, the corresponding risks were 0.67 (95% CI, ) and 0.53 (95% CI, ), respectively. Among those using HRT throughout the follow-up the risks were 0.66 (95% CI, ) for any and 0.37 (95% CI, ) for distal forearm fracture, and after adjusting as done previously 0.62 (95% CI, ) and 0.41 (95% CI, ), respectively. The crude relative risk of other than distal forearm fracture was 0.83 (95% CI, ) among part-time continuous HRT users compared with neverusers. This risk was significantly higher than that for distal forearm fracture (hazard ratio 0.49 [95% CI, ]). After adjustment for potential confounders (cf. as done previously) the relative risks of other than distal forearm fracture even for part-time and continuous users each became significant (Tables 3 and 4). Past use of HRT did not protect from fracture. The inclusion of dairy calcium intake as a continuous variable, as well as use of calcium supplements and smoking as dichotomous variables, in multivariate models did not affect the associations between HRT and. A subanalysis on those who did not use calcium supplements at baseline (n 4405) gave a fracture risk of 0.69 (95% CI, ) and a distal forearm fracture risk of 0.52 (95%

4 DOES HRT PREVENT FRACTURES? 531 TABLE 3. CRUDE RELATIVE RISKS FOR FRACTURES AMONG POSTMENOPAUSAL WOMEN (n 7217) ACCORDING TO HRT USE Any fracture Distal forearm fracture Other fracture Use of HRT (no. of women) Fractures Never (3335) Past use only (1212) ( ) ( ) ( ) Use during follow-up Part-time (1335) ( )* ( ) ( ) Continuous (1335) ( ) ( ) ( ) s analyzed with Cox regression as hazard ratios. * p ; p ; p ; p TABLE 4. ADJUSTED a RELATIVE RISKS FOR FRACTURES AMONG POSTMENOPAUSAL WOMEN (n 7217) ACCORDING TO HRT USE Any fracture Distal forearm fracture Other fracture Use of HRT (no. of women) (women) Never (3335) Past use only (1212) ( ) ( ) ( ) p Use during follow-up Part-time (1335) ( )* ( ) ( ) Continuous (1335) ( ) ( ) ( ) s analyzed with Cox regression model. a Adjusted for age, time since menopause, BMI, number chronic health disorders, and history of previous. * p ; p ; p ; p ; p ; p CI, ) for all HRT users after adjusting for age, BMI, health disorders, previous fracture, time since menopause, dairy calcium intake, and smoking. A subanalysis on those who reported the use of calcium supplements at baseline (n 2812) gave the following adjusted fracture risk estimates related to the use of HRT: 0.62 (95% CI, ) for any and 0.51 (95% CI, ) for distal forearm fracture. Subanalyses on (1) the healthy (HR 0.68 [95% CI, ]) and (2) the sick (0.70 [95% CI, ]) or on (3) due to a fall on the same level (0.60 [95% CI, ]) or (4) not including those caused by car crash or bicycle accident (0.67 [95% CI, ]) did not alter the associations between HRT and any fracture. DISCUSSION The main results of our study were that part-time and continuous but not past HRT use was associated with a diminished risk of early postmenopausal fracture in general and that of distal forearm fracture in particular. The strengths of this investigation are its large size, population-based design, and high response rate (92%). All were validated. Most of the earlier studies are carried out with old postmenopausal women as compared with our early postmenopausal study population. Previous studies mostly have been carried out with hip fracture as the endpoint, but we focused on all fracture types. A limitation was that this study was observational. Thus, confounding (18,19) is possible because HRT users were healthier and therefore might have sustained fewer. However, adjusting for health status and subanalyses on the healthy and the sick as well as on those with negative and positive fracture history revealed similar relative risks related to HRT, which precludes the possibility of a substantial confounding. Another limitation is that the information on HRT use was based on self-report, which was not validated. Therefore, if there was some indistinct information, we declared it by telephone. Recall of might have been differential according to the HRT use category leading to information (recall) bias. An observational study cannot preclude it. However, differences of fracture risks between forearm and other between past and part-time/continuous users or suggestions that HRT improves memory (20) do not lend support to information bias as the only explanation of the results. One also can raise a question about the significance of the indication of HRT use as to its effect on fracture risk. Our

5 532 RANDELL ET AL. women usually had (85%) used HRT for menopausal symptoms. Proportions of HRT use for menopausal symptoms/ total HRT use were similar in fractured and nonfractured women. Presumably, the duration of HRT use to relieve menopausal symptoms is short compared with its use for other purposes such as prevention or treatment of osteoporosis (3 7%). However, when used throughout the follow-up period of the study no differences between indications were found. During the follow-up of this study in bisphosphonates were not yet in use in the study area. Also, only very few persons (0.1% of our study subjects) used calcitonin at baseline. Thus, calcium supplements were the most frequently used osteoporosis drug besides HRT. Thirty-nine percent of our study subjects reported the use of calcium supplements at baseline. The use of calcium supplements appeared to have no effect on fracture risk nor any additive effect to that of HRT. Several things may have caused this. First, the nutritional calcium intake in this population was quite high approximately 1000 mg/day. (21) Second, the content of elemental calcium in these supplements varied considerably. Third, the use of supplements often had been started on own initiative and dosage or duration of treatment may have been irregular. Some of the may have remained unreported. According to our validation study, self-reports missed 5% of forearm and 22% of all that have been incorrectly classified as non. (22) This may have slightly diluted risk estimates. Our study confirms the earlier findings of HRT and fracture studies (16,17) that HRT has a beneficial effect on postmenopausal bone strength. We found the most protective effect for distal forearm fracture, which is the most common fracture among early postmenopausal women, with the incidence rising sharply after 50 years of age and plateauing again 15 years later. (16,23) There also was a tendency for a positive effect of HRT on other. However, there were not enough cases to confirm the finding for other specific fracture types. The mechanism of action is most likely the skeletal effect of HRT. However, recently, we have found that HRT may prevent early postmenopausal falls. (24) In addition, adjustment for bone mineral density (BMD) in our clinical trial did not decrease fracture risk related to HRT. (11) Contrary to the Danish study, (12) we found that the HRT effect remained even after excluding the high-energy (car crash). Past use of HRT did not offer any protective effect against compared with current use. These results imply that women need to use HRT continuously after menopause to minimize the risk for fracture. We conclude that HRT has a beneficial effect on prevention of, especially on that of distal forearm fracture, in early postmenopausal women. ACKNOWLEDGMENTS This study was supported by grants from the European Foundation for Osteoporosis, the Yrjö Jahnsson Foundation, the Ministry of Health and Social Affairs, and the Academy of Finland. REFERENCES 1. Christiansen C, Christiansen MS, Transbol I 1981 Bone mass in postmenopausal women after withdrawal of oestrogen/ gestagen replacement therapy. Lancet 1: Lufkin EG, Wahner HW, O Fallon WM, Hodgson SF, Kotowicz MA, Lane AW, Judd HL, Caplan RH, Riggs BL 1992 Treatment of postmenopausal osteoporosis with transdermal estrogen. Ann Intern Med 117: Lindsay R, Christiansen C, Einhorn TA, McKay Hart D, Ljunghall S, Mautalen CA, Meunier PJ, Morii H, Mundy GR, Rapado A, Stevenson J 1997 Who are candidates for prevention and treatment for osteoporosis? Consensus development statement. Osteoporos Int 7: Komulainen M, Kroger H, Tuppurainen MT, Heikkinen AM, Alhava E, Honkanen R, Jurvelin J, Saarikoski S 1999 Prevention of femoral and lumbar bone loss with hormone replacement therapy and vitamin D3 in early postmenopausal women: A population-based 5-year randomized trial. J Clin Endocrinol Metab 84: Cauley JA, Seeley DG, Ensrud K, Ettinger B, Black D, Cummings SR 1995 Estrogen replacement therapy and in older women. Study of Osteoporotic Fractures Research Group. Ann Intern Med 122: Reginster J-Y, Bruyere O, Audran M, Avouac B, Body J-J, Brandi M-L, Gennari C, Kaufman JM, Lemmel EM, Vanhaelst L, Weryha G, Devogclaer JP 2000 Do estrogens effectively prevent osteoporosis-related? Calcif Tissue Int 67: Kiel DP, Felson DT, Andersson JJ, Wilson PWF, Moskowitz MA 1987 Hip fracture and the use of estrogens in postmenopausal women. The Framingham Study. N Engl J Med 317: Naessén T, Persson I, Adami H-O, Bergström R, Bergkvist L 1990 Hormone replacement therapy and the risk for first hip fracture. Ann Intern Med 113: Kanis JA, Johnell O, Gullberg B, Allander E, Dilsen G, Gennari C, Lopes Vaz AA, Lyritis GP, Mazzuoli G, Mirauet L, Passeri M, Perez Cano R, Rapado A, Ribot C 1992 Evidence for efficacy of drugs affecting bone metabolism in preventing hip fracture. BMJ 305: Michaëlson K, Baron JA, Johnell O, Persson I, Ljunghall S 1998 Variation in the efficacy of hormone replacement therapy in the prevention of hip fracture. Osteoporos Int 8: Komulainen MH, Kröger H, Tuppurainen MT, Heikkinen A-M, Alhava E, Honkanen R, Saarikoski S 1998 HRT and Vit D in prevention of non-vertebral in postmenopausal women; a 5 year randomized trial. Maturitas 31: Mosekilde L, Beck-Nielsen H, Sorensen OH, Nielsen SP, Charles P, Vestergaard P, Hermann AP, Gram J, Hansen TB, Abrahamsen B, Ebbesen EN, Stilgren L, Jensen LB, Brot C, Hansen B, Tofteng CL, Eiken P, Kolthoff N 2000 Hormonal replacement therapy reduces forearm fracture incidence in recent postmenopausal women results of the Danish Osteoporosis Prevention Study. Maturitas 36: Hutchinson TA, Polansky SM, Feinstein AR 1979 Postmenopausal oestrogens protect against of hip and distal radius. A case-control study. Lancet 2: Weiss NS, Ure CL, Ballard JH, Williams AR, Daling JR 1980 Decreased risk of of the hip and lower forearm with postmenopausal use of estrogen. N Engl J Med 303: Williams AR, Weiss NS, Ure CL, Ballard J, Daling JR 1982 Effect of weight, smoking, and estrogen use on risk of hip and forearm in postmenopausal women. Obstet Gynecol 60: Mallmin H, Ljunghall S, Persson I, Bergström R 1994 Risk factors for of the distal forearm: A population-based case-control study. Osteoporos Int 4:

6 DOES HRT PREVENT FRACTURES? 17. Honkanen RJ, Honkanen K, Kröger H, Alhava E, Tuppurainen M, Saarikoski S 2000 Risk factors for perimenopausal distal forearm fracture. Osteoporos Int 11: Hemminki E, Sihvo S 1993 A review of postmenopausal hormone therapy recommendations: Potential for selection bias. Obstet Gynecol 82: Grodstein F 1996 Invited commentary: Can selection bias explain the cardiovascular benefits of estrogen replacement therapy? Am J Epidemiol 143: Kampen DL, Sherwin BB 1994 Estrogen use and verbal memory in healthy postmenopausal women. Obstet Gynecol 83: Kumpulainen J, Tahvonen R 1989 Report on the activities of the subnetwork on trace elements status in food. In: Report on the consultation of the FAO European Co-operative Research Network on trace elements, Lausanne, Switzerland, 5 8 September Rome: FAO. 22. Honkanen K, Honkanen R, Heikkinen L, Kröger H, Saarikoski S 1999 Validity of self-reports of in perimenopausal women. Am J Epidemiol 150: Winner SJ, Morgan CA, Evans JG 1989 Perimenopausal risk of falling and incidence of distal forearm fracture. BMJ 298: Randell KM, Honkanen RJ, Komulainen MH, Tuppurainen MT, Kröger H, Saarikoski S 2001 Hormone replacement therapy and risk of falling in early postmenopausal women a population-based study. Clin Endocrinol (Oxf) 54: Address reprint requests to: Kaisa M. Randell, M.D. Department of Obstetrics and Gynecology Kuopio University Hospital P.O. Box Kuopio, Finland Received in original form January 31, 2001; in revised form July 31, 2001; accepted September 14, 2001.