skip navigation

OffLine Winter 1999

Volume 11 Number 3

Bones, just like the rest of our bodies, are not immune to the aging process. They achieve their greatest strength (density) around age 30 and then begin to "age" or lose density. This phenomenon can result in osteoporosis, a condition characterized by fragile bones or bones that can easily be broken. Osteoporosis is the culprit in millions of fractures per year in the United States with associated health care costs of $13.8 billion. Fractures of the hip are by far the most serious, accounting for more death, disability and health care costs than all other osteoporotic fractures combined. Twelve to twenty percent of people who break their hip die in the year following fracture.

Some factors that predispose people for osteoporosis are immutable—older age, being female, a maternal history of fracture and menopause. However, researchers have identified several strategies that are likely to prevent fragile bones in older age: weight bearing physical activity, quitting smoking, and adequate consumption of calcium and Vitamin D. In addition, certain medications have been shown to have both adverse (e.g., steroids, some sedatives) and beneficial (e.g., estrogen, hormone replacement therapy) effects on bone density and strength.

The relationship between specific medications and bone density/risk fracture has been a fertile area of CHS research for the past decade. Three large randomized trials have examined the hypothesized positive effects of two medications, alendronate and thiazide diuretics, on bone density and/or fracture risk. A four-year prospective study is designed to address concerns that Depo-Provera, an injectable birth control method, may adversely affect the development of peak bone density in young women. This issue of Off Line discusses these studies and other developments in CHS osteoporotic research.

Background—Much of this research focuses exclusively on women because of their greater likelihood of developing osteoporosis compared to men. This is due to two factors: 1) Men achieve higher levels of peak bone density than women, resulting in a natural advantage when bones start aging; and 2) Women experience menopause, which due to estrogen depletion, causes an accelerated period of bone loss that may continue for some years. Long-term hormone replacement therapy (HRT) has been demonstrated to reduce a postmenopausal woman's risk for osteoporosis. CHS' research program related to HRT was covered in Off Line Volume 10 Number 2.

Bone mineral density (BMD) is the most measurable marker of bone strength. It is often measured by dual-energy X-ray absorptiometry, a radiographic, non-invasive method. Measurements are usually taken for the hip, spine, femoral neck and whole body.

In order to understand some of the studies described below, it's necessary to distinguish between "clinical fractures" and "non-clinical vertebral fractures". "Clinical fractures" (e.g., hip, wrist, some spine or vertebral) cause pain and are diagnosed by a physician. A "non-clinical vertebral fracture" is a non-painful or "silent" shortening of a vertebra occurring either spontaneously or through minimal trauma and detectable only through X-ray. About one-third of vertebral fractures cause pain and are classified as clinical. Vertebral fractures account for height loss and the dowager's hump (forward curvature of the spine, also called kyphosis) observed among some elderly people.

Fracture Intervention Trials—The Fracture Intervention Trials (FIT) were conducted at 11 sites in the U.S, including Group Health under the direction of CHS investigator Andrea LaCroix. FIT consisted of two randomized trials examining the effect of the medication alendronate on fracture incidence among postmenopausal women with low bone density at baseline. Alendronate, developed by Merk & Co. from a class of drugs called bisphophonates, is a potent medication that binds to the surface of bones and blocks resorption.

The first trial, called the vertebral fracture trial (VF) enrolled women who, according to X-rays, had a pre-existing vertebral fracture; the second trial, called the clinical fracture trial (CF) randomized women without pre-existing vertebral fractures. Why the distinction? Existence of a prior fracture demonstrates that a woman has established osteoporosis and places her at very high risk for future fractures. Women who have low bone mass but no fracture history have not yet demonstrated bone fragility and some of them never will. Alendronate had already been studied in older women with pre-existing fractures; FIT-CF was the first randomized trial to examine the effect of this medication on postmenopausal women with low bone mass but no demonstrated fragility (e.g., no history of osteoporotic fracture).

FIT-VF randomized 2027 postmenopausal women with existing vertebral fractures to receive placebo or alendronate over a three-year period. Findings that alendronate use increased bone density and decreased the risk of developing a "new" non-clinical vertebral fracture by 47 percent confirmed previous research. In addition, FIT-VF provided the first evidence from a randomized trial that alendronate reduced the risk of hip, wrist and clinical vertebral fractures by about 50 percent. Not even HRT has been shown in randomized trials to reduce the incidence of hip fractures.

FIT-CF randomized 4432 postmenopausal women with low bone mass but no pre-existing vertebral fractures to receive either placebo or alendronate over a four year period. Given that FIT-CF subjects had not yet demonstrated bone fragility, investigators hypothesized that alendronate's effects might be more pronounced in women with the lowest bone density levels. Based on an apriori decision, analyses were stratified according to tertiles of baseline femoral neck BMDs. Although all women had low bone mass relative to normative data, only those whose levels were in the lowest tertile met established criteria for osteoporosis.

FIT-CF found that four years of alendronate use increased bone density and decreased the risk of non-clinical vertebral fractures (by about 50 percent) compared to placebo. This effect held up regardless of baseline BMD tertile. However, alendronate reduced the risk of any clinical fracture only among women whose baseline bone density levels indicated that they had osteoporosis. It is unclear whether 4 more years of treatment with alendronate would have reduced the clinical fracture risk among women without osteoporosis

Taken together, the results of FIT VF and CF support the conclusion of the National Osteoporosis Foundation that the clearest benefit of alendronate is found among "high risk" postmenopausal women who have either a vertebral fracture or osteoporosis confirmed by bone densitometry.

TOPAS—Thiazide diuretics, medications commonly prescribed for hypertension, have also been hypothesized to play a beneficial role in osteoporosis prevention. The weight of the evidence from observational studies suggests that thiazide users are about one-third less likely to fracture their hip than are non-users. Because observational studies cannot establish causation, LaCroix and colleagues conducted a three year randomized trial among 320 older Group Health men and women without high blood pressure to assess the effects of low dose thiazide therapy on bone density at the hip. Subjects were randomized to one of the following 3 daily medication regimens: 1)placebo; 2)12.5 mg of thiazides; and 3)25 mg thiazides.

After 3 years. low dose thiazide therapy prevented the loss of bone mineral density at the hip compared to placebo. The 25 mg/day dose was most clearly beneficial for women while the 12.5 mg dose was most beneficial for men. A dose response relationship between thiazide dosage and change in bone density at the hip was observed for women but not men. The therapy was generally well tolerated except by men taking 25mg/day whose rate of discontinuation was 39 percent.

The magnitude of the benefit was modest—the change in BMD in the 25/mg day group relative to baseline was +0.61 percent compared to a -0.32 percent change in the placebo group over three years. However, these small effects, if accumulated over 10 to 20 years, could explain the one-third reduction in hip fracture seen in observational studies.

Commenting on the results of FIT and TOPAS, CHS investigator Andrea LaCroix says, "We are excited to contribute new scientific evidence that increases the available options for prevention of bone loss and fracture in older adults." In addition to this research focusing on prophylactic medications, LaCroix and colleagues are beginning to tackle another aspect of the prevention equation: How to best design and implement effective osteoporosis and fracture prevention programs at Group Health. The first step is to identify the target population for such prevention programs, namely those with a high likelihood of of osteoporosis or osteopenia (borderline osteoporosis). While bone density screening (e.g., X-rays) might be the best way to identify these high risk individuals, providing screening to all people over a certain age would likely be very costly.

Project OPRA is a randomized trial comparing the effectiveness of 2 prescreening written instruments designed to identify a subset of "high risk" women who would then go on to have bone density screenings. The first instrument, developed by Merck, uses 6 questions to identify women in the top 2/3 of fracture risk. The second instrument was developed by CHS and is designed to identify women in the top 15 percent of fracture risk. Women identified by either instrument as "high risk" are invited to undergo bone density testing. The instruments will be evaluated for their ability to identify women who are truly osteoporotic or osteopenic (as defined by BMD levels) and their ability to not falsely identify women as having low BMD when their levels are normal. (Women in a third randomization group will all be invited to have a bone density screening, regardless of fracture risk.) The trial will also compare the 3 randomized groups with respect to the following: 1)initiation of HRT or other anti-osteoporosis therapies; 2)changes in other modifiable fracture risk factors; and 3) awareness of and attitudes toward osteoporosis and fracture prevention. It is hoped that the OPRA trial will provide data that can help design the most effective prevention programs for osteoporosis in managed care organizations.

Depo-Provera—Depo Provera (DP) is an injectable contraceptive estimated to have been used by more than 15 million women worldwide over the last thirty years. Following FDA approval in 1992, a lingering safety issue concerns DP possible adverse effect on bone density in premenopausal women which might place them at future risk for osteoporosis. The basis for concern is the DP works by suppressing the production of estrogen, a hormone critical to attainment of peak bone mass. CHS investigator Delia Scholes is conducting a four-year prospective study of the relationship between DP use and bone mineral density among a cohort of reproductive age women. This study examines osteoporosis prevention from a different angle than the alendronate and thiazide trials, focusing on the attainment and maintenance of peak bone mass in younger women rather than the prevention of bone loss in the elderly.

The study has enrolled 457 female Group Health enrollees aged 18–39, 183 women who use DP for contraception and 274 who do not. Cross-sectional data showed that DP users had spine BMD levels about 2.5 percent lower than those for non-users, with the largest differences (9.4 percent) occurring in the youngest age group (ages 18–21 years). Researchers found a dose response relationship between longer DP use and decrease in bone density levels in this youngest group. At baseline, there were no significant differences in mean bone density levels for all other age groups (22–39); nor did these age groups show any evidence of a dose response relation between duration of DP use and BMD levels.

Scholes and colleagues have also begun to examine the prospective data. In the 2 years since baseline, the change in BMD in the non-user group relative to baseline was +0.44 percent compared to a -1.9 percent to -2.8 percent change in the DP user group. (The mean  percent change among DP users depended on whether they were prevalent or new users upon entry into the study). Unlike the cross-sectional analysis of bone density levels, these significant prospective findings regarding bone changes in users vs. non-users were not restricted to the youngest age group. If these trends continue, Scholes concludes that DP use may increase a woman's future risk by 50 percent. However, a major aim of the prospective analysis will be to assess the effects on bone following DP discontinuation. Preliminary results indicate that the downward trend is reversed upon stopping use of DP.

Top