Osteoporosis 2002: Headline News. (Featured CME Topic: Osteoporosis).

ALTHOUGH the title is "Headline News," this material might be more correctly classified as breaking news, since much of it has not been published. A number of items in the diagnosis of osteoporosis could have been chosen, but I focus on a particular area and present headline news or breaking news on treatment of osteoporosis and address the following 5 issues:

How fast does treatment work (when the unit of measure for the effectiveness of treatment is reduction in fracture incidence)?

How long should we treat? With treatment, how long do we see reduction in the number of fractures?

What is the evidence for the effectiveness of bisphosphonates in reducing the incidence of hip fractures?

What is the evidence for once weekly dosing of bisphosphonates?

What is the role of vertebroplasty and kyphoplasty, new procedures for the management of painful vertebral fractures?

We know that osteoporosis takes years or even decades to develop. We think that it may take months or even years of therapy to reduce the risk of fracture. The Food and Drug Administration (FDA) requires trials that show reduction in the incidence of vertebral fractures over 3 years. The question that we will try to answer is, "Could treatment reduce the incidence of fracture at an earlier time (sooner than 3 years), and if so, how much earlier?" The importance of this question is intuitively obvious. If a patient has limited life expectancy, and it takes 3 years of treatment to show a reduction in fracture, then it is probably not appropriate to treat. If something will effectively reduce the risk of fracture in 6 or 12 months in that same patient with limited life expectancy who is at risk for an osteoporotic fracture, however, who is dealing with a variety of physical, emotional, and social challenges, then it may make sense to treat; adding a fracture to the burden of her other medical and social problems may be undesirable.

The only data sets that allow an answer to this question from a prospective design are the Vertebral Effectiveness of Risedronate Therapy (VERT) studies. There were 2 pivotal trials: 1 was done in North America, (1) and the other was a multinational trial. (2) To be enrolled in the studies, patients had to either have 2 or more vertebral fractures or, in the North American trial, 1 vertebral fracture plus a low T-score. The average age of patients was late 60s to early 70s. The mean number of vertebral fractures was 2 in the North American trial and 4 in the multinational study. In contrast to the other pivotal trials of osteoporosis therapy, in which x-ray films were done at 3 years and sometimes at 2 years, subjects in these trials had yearly spine radiographs, which allowed us to answer the question, "Does treatment reduce the risk of radiographically detected spine fracture at 1 year?" The duration of these trials was 3 years. After 3 years, incidence of vertebral fracture was significantly reduced in bot h trials; there was a 41% reduction in the North American study, and 49% reduction in the multinational study. The question then is, "What happens at 1 year?" The answer is significant reduction in the incidence of fractures. There was a 65% reduction in the incidence of new vertebral fractures after 1 year of treatment in the North American trial, and 61% reduction after 1 year of treatment in the multinational trial. (3) Both of these results were highly statistically significant.

This finding prompted a re-evaluation of the data from the other large trials. While radiographs cannot be done once the trial is over, there may be other ways to get similar information from existing data sets. One large data set is the Fracture Intervention Trial (FIT) using alendronate, which included almost 6,500 women with low femoral-neck bone density. The original design had 2 arms: a vertebral-fracture arm (subjects with prevalent vertebral fractures) (4) and a clinical-fracture arm (subjects without prevalent vertebral fractures, but femoral neck T-scores of --2 or below) . (5) The data from these 2 arms were pooled and reanalyzed; it was recognized that some of the subjects in the second arm of FIT really did not have osteoporosis, which is complicated to explain; however, only about a third of the subjects in that arm had femoral-neck T-scores --2.5 or below by current standards; the analysis described here consists of all the subjects who were in the first arm of FIT and those in the second arm of FIT who had femoral-neck T-scores of --2.5 or below. In both of these trials, everyone was given calcium and vitamin D supplementation, if needed. Subjects were treated with either alendronate or placebo. Clinical vertebral fractures were reported as adverse events. This was not a primary endpoint of the trial, but all clinical fracture data were collected as a safety measure, so we can look at the time at which clinical fractures were significantly reduced by treatment. As early as 1 year after beginning treatment, there was a significant 59% reduction in clinical vertebral fractures in the group treated with alendronate compared with the group receiving the placebo. (6)

Included in the large study with raloxifene, the Multiple Outcomes of Raloxifene Evaluation (MORE) trial, were 7,705 women, average age late 60s and about 18 to 20 years postmenopausal, with low bone density. About 20% had at least 1 fracture at baseline, and another 17% had 2 or more fractures; the analysis here consists of the women with and without fractures. (7) Again, radiographs were done at 2 and 3 years, but clinical fracture data were collected throughout the trial for safety purposes. This analysis looks at clinical vertebral-fracture data with raloxifene therapy. After 1 year of treatment with raloxifene, there was a 68% reduction in the incidence of clinical vertebral fracture in patients receiving raloxifene compared with those taking a placebo. (8)

Realizing that these are different studies in different populations, we can summarize as follows: after 1 year of treatment with raloxifene, there was a 68% reduction in the incidence of clinical vertebral fractures; after 1 year of alendronate therapy, a 59% reduction; and, with risedronate treatment, the incidence of clinical vertebral fractures was reduced at 1 year by 69%. Looking again at the data on risedronate therapy for clinical vertebral fractures, it is apparent that the incidence of fractures was reduced, not only at 12 months, but also as early as 6 months. The numbers are small at 3 months, but the curve separates at that first time point. It appears that the agents that we currently have--at least raloxifene and the 2 bisphosphonates, alendronate and risedronate-- have a rapid onset of effectiveness in reducing the risk of new vertebral fractures.

These agents are effective in reducing the incidence of fracture after 6 to 12 months of treatment. Once treatment is started, we should periodically reassess the situation and re-evaluate what is known about treatment. A second question is, "How long should we treat?" There is not a clear answer to this question. One might treat for a time, withdraw treatment, and continue to monitor the patients to see if and when the antifracture effect disappears. To have evidence to support that approach, a parallel placebo group would be needed to determine when the fracture rate in the treated group came in line with the fracture rate in the placebo group. No one has done such a trial, and apparently no such trial is contemplated.

It may shed some light on this to see what happens if treatment continues after 3 years, and compare treated patients with a group continuing to take a placebo. Is there a continued reduction in fracture if the patient remains on treatment? The pivotal alendronate studies do not have data on fractures, but do give information on bone density to address this question. The original alendronate trial had 4 groups: placebo; alendronate 5 mg/day; alendronate 10 mg/day; or alendronate 20 mg/day in the first 2 years, changed to 5 mg/day in the third year. This trial has been extended 3 times, and data have been published or presented for 2 of those extensions. In the 5-year data (a 2-year extension of the 3-year trial), the subjects taking a placebo were changed to 10 mg/day of alendronate, the subjects receiving 5 mg/day stayed on the same treatment, those taking 10 mg/day stayed on the same treatment, and subjects who had been taking 20 mg/day then changed to 5 mg/day stayed on the same dose. In the final publishe d 7-year data (a second 2-year extension), patients who were in the placebo group were dismissed from the trial, patients who had been taking 5 or 10 mg/day of alendronate were kept on their originally assigned treatment, and patients who had a shift in treatment (20 mg/day for 2 years, then 5 mg/day for 3 years) were shifted a third time to receive placebo. We can look at this placebo group to tell what happens to bone density or biochemical markers of bone turnover when treatment is stopped, but we cannot look at this to tell anything about antifracture effectiveness.

Over the course of 7 years of treatment with alendronate, there was a good gain in bone mineral density (BMD) in the first year, but there was continued increase, albeit at a slower rate, throughout the entire 7-year study. At the end of 7 years of treatment, there was an average 11% increase in bone density in the lumbar spine. (9) Patients who were treated with 20 mg/day for the first 2 years, 5 mg/day through year 5, and then changed to placebo, did not gain or lose BMD when the drug was stopped (at least for the first 2 years after treatment). There has been some thought that a reservoir of drug might be left in the skeleton that might have some sustained effect. In this group of patients, biochemical markers of bone turnover began to increase within 3 months of the time the drug was stopped. Two years after stopping the drug, they were still below their baseline values from the start of the study, so it is not a complete reversal of effect. It is likely that some time after stopping the drug, the effects on bone density and biochemical markers of bone turnover will disappear, and that depends in part on the duration of treatment.

We do not know what will happen to fracture rates in these patients who continue treatment. If there is an increase (or even stability) in bone density, there is no guarantee that fracture rates will remain low. Two studies have addressed this by extending the study past the original 3-year period and keeping cohorts in their originally assigned groups (active treatment vs control).

The MORE Study was extended through 4 years, with the patients kept in their original groups. Continued treatment with raloxifene through year 4 provided continuing protection against vertebral fractures, both in the fourth year and over the course of 4 years. There was a significant reduction in the incidence of fractures in the 2 subgroups. (10) In the first 3 years, women who had prevalent fractures on entry into the trial had 30% reduction in fracture incidence, and women in the subgroup without prevalent fractures had a 55% reduction in the number of fractures. In these same 2 cohorts in the fourth year, as a stand-alone analysis, there was a 38% reduction in the incidence of fractures in the group with prevalent fractures and a 50% reduction in the group without prevalent fractures. So, the answer to the question of whether raloxifene has a continued antifracture effect in year 4 and through year 4 is yes.

As an aside, this trial provides some interesting information with regard to the clinical diagnosis of vertebral fracture. It may be optimistic to say that only about 1 of 3 fractures that we see on x-ray film is diagnosed as a clinical event. In the real world, it may be more like 1 of 4, or maybe 1 of 5. The participants in this trial were known to have osteoporosis, had regular visits to talk with a study nurse who was gathering information about their osteoporosis, and were measured with a stadiometer at every visit to determine changes in height. Fractures were ascertained either from clinical reports (at the time of the event) or from vertebral radiographs. Over the first 2 years, there was a gradual increase in the number of new clinical fractures, with a total of 2.5% to 3.0% incidence of clinical vertebral fractures over 2 years. When the first follow-up x-ray films were done at 2 years, the incidence jumped to about 8% (total of clinical plus radiographic vertebral fractures). In patients with osteo porosis who have new back pain or show some height loss, therefore, there should be a high index of suspicion that something may have happened in the spine.

The risedronate multinational trial was extended through 5 years, and in that 2-year extension, the subjects remained in their originally assigned groups--a placebo group or a risedronate treatment group. In the extension period (years 4 and 5), there was a significant reduction in incidence of vertebral fractures with risedronate treatment and a trend, though not statistically significant, for a reduction in the incidence of nonvertebral fractures. (11) In years 4 and 5 of the study, the patients who were treated with risedronate had half as many vertebral fractures as patients who were taking placebo. There was a trend toward risk reduction for nonvertebral fractures over the first 3 years, a trend toward risk reduction over the extension period, and significant reduction over 5 years with risedronate treatment.

We are supposed to practice evidence-based medicine. Evidence indicates that all of the drugs we currently use for treatment of osteoporosis reduce the incidence of vertebral fracture. What are the relative strengths of the data for reduction in the incidence of vertebral fracture? First, has antifracture effectiveness been shown in more than 1 study or in more than 1 stratum? The effectiveness of alendronate and risecironate for reducing the risk of vertebral fractures has been shown in multiple trials, and the effectiveness of raloxifene has been shown in different strata (with and without prevalent vertebral fractures). Second, has the incidence of radiographic vertebral fractures been reduced after 1 year of treatment? The answer is yes for risedronate. Third, do data show that the incidence of clinical vertebral fractures is reduced as early as 12 months after treatment? The answer is yes for alendronate, raloxifene, and risedronate (risedronate has also been shown to reduce the incidence of fractures af ter 6 months). Fourth, is the incidence of radiographic deformities reduced through 4 or 5 years and in a fourth and/or fifth year of treatment? This is tricky. The Clinical Fracture Arm of FIT was a 41/2-year study, (5) but it did not show antifracture effectiveness during the fourth year. The Prevent Recurrence of Osteoporotic Fractures trial using calcitonin was a 5-year study, but it does not show the effectiveness of treatment in that last time period. Raloxifene has been shown to reduce the incidence of fractures in a fourth-year extension period, and risedronate has been shown to reduce fracture incidence in 4- and 5-year extension periods.

We know that spinal fractures result in a significant burden, but most people would agree that hip fracture is the most serious consequence of osteoporosis. Bisphosphonates are the only class of agents that have data showing reduction in the incidence of hip fractures and other nonvertebral fractures. The first study that showed a reduction in hip fracture incidence with any treatment for osteoporosis was the FIT study (vertebral-fracture arm) with alendronate, (4) and 3 other trials of alendronate in which hip-fracture data were collected. (12-14) Overall, there were 77 hip fractures across these 4 studies. A meta-analysis of these trials shows that the relative risk for hip fracture was 0.48, a 52% reduction.

The only study done thus far in which hip fracture was the primary endpoint is the Hip Intervention Program (HIP) study using risedronate. (15) The original plan was to enroll women aged 70 to 79 years who had low bone density, T-scores of -4 or below, or T-scores of -3 plus a risk factor (women with osteoporosis, group 1). It is amazing that the investigators could get almost 5,500 women who met these criteria to enroll in a 3-year trial. Despite that large number, enrollment was not thought to be adequate, so a second pathway for entry into the trial was added: women who were believed to be at high risk of hip fracture because of their age (80 years or older) and the presence of clinical risk factors (group 2). The risk factors were not so much risk factors for osteoporosis as they were risk factors for fragility or falling, such as impaired psychomotor skills, difficulty standing, and poor tandem gait. Most of the subjects in group 2 did not have bone density measurements, so it is not known whether they h ad osteoporosis.

In the primary analysis, risedronate therapy significantly reduced the incidence of hip fracture by 30% over the 3 years of the trial. It was planned in advance to analyze the results of the 2 groups separately. About 5.5% of group 2 subjects who received placebo had fractures over 3 years, compared with 3% of the group 1 patients who received placebo. This shows that the risk of fracture was high in both groups. In group 1 (women with osteoporosis), treatment resulted in a significant 40% reduction in hip fracture incidence. In group 2 (the women who were thought to be at high risk for fracture due to frailty or falling), there was no significant effect of treatment. The message here is twofold: (1) patients with osteoporosis, probably regardless of age, will benefit from treatment with a bisphosphonate in terms of reduction in hip fracture risk; and (2) a patient who is likely to fall, has poor balance, poor motor control, or poor vision, should have those nonskeletal risk factors addressed. Treating patien ts with an osteoporosis drug will not change their likelihood of falling, improve their balance or eyesight, and will not protect them in the event of a fall. Thus, in evaluating patients, think of skeletal risk factors (low bone density, previous fractures) as risk factors for fracture that should lead to treatment with an osteoporosis drug, but think also about nonskeletal risk factors for fractures, such as frailty and poor balance, that need a different kind of intervention.

If their risk of falling cannot be changed, patients can be protected from injury if they do fall by putting them in hip protectors. One study showed a 52% reduction in hip-fracture risk in nursing-home patients who were assigned to wear hip protectors, compared with a control group. (16) Most of the subjects who had hip fractures in the group that was assigned to wear hip protectors were not wearing the hip protector when they fell and fractured.

Bisphosphonates are known to be effective treatment for osteoporosis, but the daily dosing can be confining. One patient was told that the bisphosphonate must be taken first thing in the morning on an empty stomach, then a 30-minute wait before eating. She responded, "I have a life! I cannot do that." Once-a-week dosing is certainly more convenient and appears to be equally effective. It would be good to have fracture data for once-a-week dosing, but the studies looked at BMD and biochemical markers of bone turnover. Alendronate is already approved for once-a-week dosing. Data that establish this are from a 1-year, double-blind, randomized study (17) that was extended through 2 years (the 2-year results have just been submitted for publication). This trial included over 1,200 women with osteoporosis who received 1 of 3 alendronate regimens: 10 mg daily, 35 mg twice weekly, or 70 mg once weekly. Everyone in the trial received calcium and vitamin D supplements. Each subject took the same number of pills, regard less of regimen. Women in the 10 mg/day group took the standard dose (10 mg/day), in 7 active pills, plus 3 placebo tablets. Women in the group taking 35 mg twice a week took 2 active pills each week, plus 8 placebo tablets. The women who were assigned to get 70 mg once a week took 1 active pill and 9 placebo tablets throughout the week. Bone mineral density was the primary endpoint. There were essentially no differences at 12 months between the 3 groups in BMD at the lumbar spine and the total hip. These postmenopausal women had elevated bone-resorption markers when they entered the trial; treatment resulted in a decrease in bone resorption into the range of about the premenopausal mean, with almost identical effects for the 3 different dosing regimens.

A similar study (multicenter, randomized, double-blind trial of about 1,200 women over age 50 years with osteoporosis) looked at the standard regimen of risedronate (5 mg/day) versus weekly dosing of risedronate at 35 mg or 50 mg. There were similar changes in BMD at the spine and hip for all 3 regimens. (18) An application for approval of the once-a-week dose of 35 mg of risedronate was filed with the FDA in June 2001.

Parathyroid hormone (PTH) treatment for osteoporosis is headline news. The FDA has sent an "approvable" letter to Eli Lilly and Company. The sponsor and the agency are working out the language that will appear in the package insert and some of the technical issues in delivery of a drug that will be given by daily subcutaneous injections. Physicians who have not followed the literature may wonder why we are discussing PTH treatment for osteoporosis, since in medical school they were taught that too much PTH has a negative effect on the bone. That is true if the exposure to PTH is continuous. If rats are given continuous PTH, there is a dramatic stimulation of bone resorption that is bad for bone. Rats given a subcutaneous injection of PTH show little stimulation of osteoclasts but have enhanced bone formation. Exposure to PTH for an hour results in an even more striking increase in bone formation, thus there seems to be a "magic" time of exposure to PTH over the course of a day that dissociates the negative ef fect on bone resorption from the positive effect on bone formation and allows an anabolic action of PTH on the skeleton.

The pivotal trial with PTH was a large one, but two others were published before that one. One study was done in women who had osteoporosis, with either a T-score of -2.5 or below or a fragility fracture. All had been taking hormone replacement therapy (HRT) for an average of about 9 years. They continued to take HRT and were given either PTH or placebo injections in a random design. The bone density of the women taking estrogen remained stable over the 3 years of this trial. That does not mean estrogen does not cause an increase in bone mass. These women had been taking estrogen for 9 years, and they had probably reached a plateau in terms of any gains they might have had. The women who continued taking estrogen with added PTH showed a dramatic increase in bone mass, almost 15% after 3 years (this compares with 6% to 8% after 3 years' treatment with bisphosphonates), and a significant (though lesser) increase in bone mass in the hip. (19)

The pivotal study of PTH was presented in 2000 at the Endocrine Society and published last year. (20) This included more than 1,600 women with at least 1 vertebral fracture. Everyone in the trial received calcium and vitamin D supplements, then were given either placebo or 1 of 2 different doses of PTH -- 20 [micro]g or 40 [micro]g--by daily subcutaneous injection. While this was planned to be a 3-year trial, it was stopped early because of events outside the trial; thus, the median and mean follow-up times were 21 months and 19 months, respectively. Information herein is based on data on bone density and fractures in a trial lasting 1[1/2] to 2 years. Bone density increased dramatically, almost 10% in the spine over 2 years with the 20 [micro]g dose of PTH, and almost 14% in the spine with the 40 [micro]g dose. There were lesser but significant increases of 3% to 5% in the femoral neck and the total hip, slightly greater with the 40 [micro]g dose than with the 20 [micro]g dose. The main endpoint in osteoporo sis treatment trials is reduction of fracture incidence, not bone density. While subjects receiving the higher dose of PTH showed a greater increase in bone density, both doses were effective in reducing the incidence of fractures and their effects were similar. The reduction in the incidence of new vertebral fractures with PTH treatment was 65% in patients receiving 20 [micro]g/day and 69% in those receiving 40 [micro]g/day. The effect on the incidence of non-vertebral fractures was about 50% to 55% reduction with the 2 different doses of PTH. The group receiving 20 [micro]g of PTH had few adverse events; subjects taking the higher dose of PTH did show a significant increase in the incidence of nausea and headaches. Serum calcium level was elevated at least once in 9% of patients in the group receiving 20 [micro]g, but only 3 had repeated elevations of serum calcium level, thus hypocalcaemia and hypercalciuria really do not seem to be problems with subjects taking the 20 [micro]g dose, and the 20 [micro]g do se has been submitted for approval by the FDA.

The last topic in the news today is vertebroplasty and kyphoplasty. Both of these procedures involve percutaneous injection of bone cement. Vertebroplasty was started in France in 1984 and in the United States in 1995. In 1997, Dr. Jacques Dion, head of interventional radiology at Emory University in Atlanta, was the first person to do this procedure in North America when he was at the University of Virginia. The procedure uses an 11-gauge needle inserted into the substance of the vertebral body, and bone cement is then injected through the needle. The idea of the procedure is to relieve pain, which it appears to do, but the mechanism is not clear. There have been no randomized trials of placebo versus procedure or sham procedure versus real procedure. There are at least 2 possible mechanisms by which this procedure could reduce pain. One is that when bone is cemented, the bone is stabilized, thereby relieving pain. Another possibility has to do with the heat that is produced as the bone cement is introduced into the vertebral body. It may be that the increase in temperature within the vertebral body causes damage to the nerve endings, thereby relieving pain. Vertebroplasty, while it may relieve pain, would not be expected to reverse deformity.

Kyphoplasty involves a preliminary step of inflating a balloon within the vertebra, expanding the collapsed vertebra, then withdrawing the balloon and filling the space with bone cement. A report recently showed that some of these vertebral deformities are actually reversible or unstable, however. (21) If the patient is seated, a crush deformity can be seen; however, if an x-ray film is obtained with the spine hyperextended, positioning can actually reverse some of the deformity. Whether kyphoplasty is effective in reversing deformity is not clear. The developers of this procedure planned to do a randomized trial, but enrollment has been disappointingly slow. There are no controlled trials of kyphoplasty in the literature. There is a series of case reports that I reviewed and published with colleagues late last year. (22) In the series, pain relief after the procedure was reported to be between 60% and almost 100% of the time, but it is not clear whether this is a placebo effect, due to heat as the bone-cemen t cures, or due to stabilization of the bone fragments. There are short-term complications, in that bone cement may extrude and put pressure either on the nerve roots or the spinal cord. In several of these small series, 1 or 2 patients needed emergency decompressive surgery.

A particular concern is the increased risk of fracture in adjacent vertebrae. We know that people who have already had 1 vertebral fracture are at high risk for having another. One study of vertebroplasty, which consisted of about 80 patients followed for 4 years, showed that the patients were at high risk for fracture, but the likelihood that they would fracture a vertebra adjacent to one that had been treated was significantly higher than the likelihood of fracture in vertebrae that were remote from the level that was treated with the procedure. (23) When I was at Emory University, I had a patient who called me complaining of severe pain from a vertebral fracture. I made arrangements for her to have a vertebroplasty in L4. I saw her again 6 months later, and what I had forgotten is that now that she had the telephone number for the interventional radiology department, so while I had not heard from her in the intervening 6 months, they had. She had had 6 vertebroplasties, about 1 per month, going up the spin e from the first one in L4. Thus, when sending patients for this procedure, make sure that they are being treated for their underlying osteoporosis and that they know who to call if they have continuing problems. Most patients who have painful vertebral fractures do not get diagnosed in the first place, and get over their pain within a period of a few weeks. If the pain lasts more than 2 or 3 months, or if it is disabling, then it is appropriate to consider either vertebroplasty or kyphoplasty. The relative safety of these 2 procedures is not clear. Intuitively, vertebroplasty might seem safer, because kyphoplasty involves an extra step. It may be the opposite, though, because vertebroplasty uses pressure to get the bone cement into the bone, whereas with kyphoplasty, space has been made for the bone cement and it is perhaps more controlled. Head-to-head trials of these 2 procedures for safety as well as short-term and long-term efficacy are needed.

References

(1.) Harris ST, Watts NB, Genant HK, et al: Effects of risedronate treatment on vertebral and nonvertebral fractures in women with postmenopausal osteoporosis--a randomized controlled trial. JAMA 1999; 282:1344-1352

(2.) Reginster JY: Risedronate increases bone mineral density and reduces the vertebral fracture incidence in postmenopausal women. Clin Exp Rheumatol 2001; 19:121-122

(3.) Watts N, Roux C, Genant H, et al: Risedronate reduces vertebral fracture risk after the first year of treatment in postmenopausal women with established osteoporosis (Abstract). J Bone Miner Res 1999; 14(suppl 1):S136

(4.) Black DM, Cummings SR, Karpf DB, et al: Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet 1996; 348:1535-1541

(5.) Cummings SR, Black DM, Thompson DE, et al: Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures-- results from the fracture intervention trial. JAMA 1998; 280:2077-2082

(6.) Black DM, Thompson DE, Bauer DC, et al: Fracture risk reduction with alendronate in women with osteoporosis: the Fracture Intervention Trial. J Clin Endocrinol Metab 2000; 85:4118-4124

(7.) Ettinger B, Black DM, Mitlak BH, et al: Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene--results from a 3-year randomized clinical trial. JAMA 1999; 282:637-645

(8.) Maricic M, Adachi J, Meunier P, et al: Raloxifene 60 mg/day has effects within 12 month in posmenopausal osteoporosis treatment and prevention studies (Abstract). Arthritis Rheum 2002; 43(suppl):S197

(9.) Tonino RP, Meunier PJ, Emkey R, et al: Skeletal benefits of alendronate: 7-year treatment of postmenopausal osteoporotic women. J Clin Endocrinol Metab 2000; 85:3109-3115

(10.) Eastell R, Adachi J, Harper K, et al: The effects of raloxifene on incident vertebral fractures in postmenopausal women with osteoporosis: 4-year results from the MORE trial (Abstract). J Bone Miner Res 2000; 15(suppl 1):S229

(11.) Watts NB, Brown J, Hosking D, et al: Sustained fracture risk reduction over 5 years with risedronate therapy (Abstract). J Bone Miner Res 2001; 16(suppl 1):S212

(12.) Cummings SR. Black DM, Thompson DE: Alendronate reduces the risk of vertebral fractures in women without pre-existing vertebral fractures: results of the Fracture Intervention Trial (Abstract). J Bone Miner Res 1997; 12(suppl):S149

(13.) Karpf DB, Shapiro DR. Seeman E, et al: Prevention of non-vertebral fractures by alendronate--a meta-analysis. JAMA 1997; 277:1159-1164

(14.) Pols HAP, Felsenberg D, Hanley DA, et al: Multinational, placebo-controlled, randomized trial of the effects of alendronate on bone density and fracture risk in postmenopausal women with low bone mass: results of the FOSIT study. Osteoporos Int 1999; 9:461-468

(15.) McClung MR. Geusens P. Miller PD, et al: Effect of risedronate on the risk of hip fracture in elderly women. N Engl J Med 2001; 344:333-340

(16.) Kannus P. Parkkari J, Niemi S, et al: Prevention of hip fracture in elderly people with use of a hip protector. N Engl J Med 2000; 343:1506-1513

(17.) Schnitzer TJ, Bone FIG, Crepaldi G, et al: Alendronate 70 mg once weekly is therapeutically equivalent to alendronate 10 mg daily for treatment of postmenopausal osteoporosis. Aging Clin Exp Res 2000; 12:1-12

(18.) Lindsay R, Kendler D, McClung MR. et al: Risedronate 35 mg once a week is as effective as 5 mg daily in postmenopausal women (Abstract). Arthritis Rheum 2001; 44(suppl):S153

(19.) Cosman F, Nieves J, woelfert L, et al: Parathyroid hormone added to established hormone therapy: effects on vertebral fracture and maintenance of bone mass after parathyroid hormone withdrawal. J Bone Miner Res 2001; 16:925-931

(20.) Neer RM, Arnaud CD, Zanchetta JR, et al: Effect of parathyroid hormone (1-34) on fractures and bone mineral density in postmenopausal women with osteoporosis. N Engl J Med 2001; 344:1434-1441

(21.) Faciszewski T, McKiernan F: Classification of vertebral compression fractures: a consensus for comparison of treatment and outcome. J Bone Miner Res 2002; 17:185-191

(22.) Watts NB, Harris ST, Genant HK: Treatment of painful osteoporotic vertebral fractures with percutaneous vertebroplasty or kyphoplasty. Osteoporos Int 2001; 12:429-437

(23.) Grados F, Depriester C, Cayrolle G, et al: Long-term observations of vertebral osteoporotic fractures treated by percutaneous vertebroplasty. Rheumatology 2000; 39:1410-1414

From the University of Cincinnati Bone Health and Osteoporosis Center, Cincinnati, Ohio.

Presented at the Fifth Annual Southern Medical Association Conference on Osteoporosis, Amelia Island, Fla, February 21-24, 2002.

Reprint requests to Nelson B. Watts, MD, University of Cincinnati Bone Health and Osteoporosis Center, 222 Piedmont Ave, Suite 4300, Cincinnati, OH 45219.