The summum bonum of osteoporosis treatment is the prevention of new fractures. In vitro studies have shown that 60-80% of the compressive strength of bone is caused by its mineral content.(1) Correspondingly, observational studies show that baseline bone mineral density (BMD) measurements at multiple skeletal sites can predict various types of osteoporotic fractures in postmenopausal women.(2) In general, a decrease in BMD of 1.0 SD increases the risk of future fractures by ∼2.0-fold. The observational data from epidemiological studies have been confirmed by results of randomized clinical trials (RCTs) showing that antiresorptive agents both increase BMD and decrease vertebral fractures.(3) These considerations have led to the widespread belief that changes in BMD can serve as a surrogate for assessing treatment effects on fracture risk. Based mainly on the failure of large increases in BMD induced by sodium fluoride therapy to reduce fracture risk,(4) the U.S. Food and Drug Administration has not allowed surrogate markers for fractures such as BMD to be used as primary endpoints for assessing efficacy of antiosteoporosis drugs. However, sodium fluoride therapy alters bone crystalline structure and reduces bone strength,(5) so these results cannot be generalized to other treatments.

Raloxifene is a selective estrogen receptor modulator (SERM) with a spectrum of tissue-specific agonist-antagonist effects on estrogen target tissues but that acts on bone as an estrogen agonist.(6) In the large MORE trial of 7705 women with osteoporosis,(7) 3 years of raloxifene treatment in dosages of 30 mg/day or 60 mg/day increased BMD at the lumbar spine by 2.6% and 2.7%, respectively, and at the femoral neck by 2.1% and 2.4%, respectively, as compared with placebo. These modest increases in BMD were associated with large decreases in vertebral fractures of 38% and 41% for the two dosages, respectively. Similarly, in the PROOF trial, treatment with nasal spray calcitonin increased lumbar spine BMD by only 1.2% but decreased vertebral fracture risk by 36%.(8) However, neither raloxifene(7) nor nasal spray calcitonin(8) treatment reduced the risk for hip and other nonvertebral fractures. By contrast, the more potent antiresorptive agents estrogen,(9) alendronate,(10) and risedronate(11) increased lumbar spine BMD over 1-3 years by 5.3-8.8%, but RCTs with these agents have resulted in reductions in the vertebral fracture rate of about 40-50% (reviewed by Faulkner(12)), only slightly more than were achieved with raloxifene or calcitonin therapy.

In this issue of the Journal, Sarkar et al.(13) report on an extensive statistical reanalysis of the MORE data to assess further the relationship between changes in BMD and changes in vertebral fracture risk. Using logistic regression models with the two dosages of raloxifene pooled, they find that fracture risk varied inversely with baseline BMD values but that the decrease in vertebral fracture risk was similar across the range of increases in femoral neck or lumbar spine BMD and that only a small fraction of the observed reduction in vertebral fracture risk with raloxifene therapy could be accounted for by the increase in lumbar spine BMD. The authors conclude that antiresorptive agents that induce greater increases in BMD cannot necessarily be assumed more efficacious in reducing fracture risk than those that elicit lesser increases. However, the article leaves unresolved the reason for discrepancy between the small increases in BMD and the relatively large decreases in vertebral fracture risk with raloxifene therapy.

Thus, there are two paradoxes that require explanation. First, how can the modest increase in BMD induced by raloxifene or nasal spray calcitonin therapy decrease fracture risk by almost the same extent as more potent antiresorptive agents that increase BMD by a 2- to 3-fold greater extent? Second, this paradox is subsumed into an even larger paradox: Why do the small increases in BMD observed after antiresorptive therapy result in a much larger decrease in vertebral fracture rate than predicted from the relationship between BMD and vertebral fracture risk found in observational studies? We will examine the second paradox first.

In 1996,(14) we advanced the hypothesis that antiresorptive agents were able to reduce vertebral fracture rates substantially without inducing large increases in BMD because of their ability to normalize high bone turnover in patients with osteoporosis. We further suggested that high bone turnover contributes substantially to vertebral fracture risk because of its disruptive effect on the microarchitecture of cancellous bone. Using histomorphometry, Parfitt et al.(15) have shown that increased osteoclastic activity associated with high bone turnover in postmenopausal women leads to perforative resorption of cancellous plates, loss of trabeculae, and trabecular discontinuity. Because much of the strength of cancellous bone is caused by its microarchitecture, the loss of trabecular connectivity and other adverse structural consequences of increased osteoclastic activity clearly could predispose to fractures of the vertebrae and other skeletal sites that have a high content of cancellous bone.

Our hypothesis was based on a reanalysis(14) of a 1-year RCT of transdermal estrogen and placebo therapy of postmenopausal women with osteoporosis(9) using computer generated, three-dimensional graphic plots that related the observed vertebral fracture rate to lumbar spine BMD and to bone turnover assessed directly by tetracycline-based histomorphometry of iliac biopsy samples. As shown in Fig. 1A, placebo-treated women with osteoporosis from this RCT had two vertebral fracture peaks, a somewhat higher peak that was associated with high bone turnover and a somewhat lower one that was associated with low baseline lumbar spine BMD. However, in the estrogen-treated women (Fig. 1B), the fracture peak associated with high turnover was not present whereas the fracture peak associated with low BMD was maintained, presumably because the increases in BMD induced by treatment were modest. Although these data plots are only semiquantitative, Fig. 1B suggests that more than one-half of the incident vertebral fractures in untreated postmenopausal osteoporotic women may be caused by high bone turnover and the remainder are caused by low BMD. Figure 1B further suggests that the short-term reduction in vertebral fractures induced by estrogen therapy is caused by mainly elimination of the fracture peak associated with high bone turnover. Because the average reduction in vertebral fracture risk produced by effective antiresorptive agents also is about 50%,(12) it is likely that the same mechanism is occurring in them as well. Moreover, the hypothesis that antiresorptive agents reduce fracture risk mainly by reducing bone turnover is supported by prospective studies(16, 17) showing that the baseline level of bone turnover markers can predict fracture risk independently of BMD. Finally, the hypothesis explains why less potent antiresorptive drugs such as raloxifene(7) and calcitonin(8) can reduce substantially vertebral fracture risk but not hip fracture risk whereas alendronate(18) and risedronate(11, 19) can reduce both vertebral and hip fracture risk. High bone turnover does weaken cortical bone by increasing its porosity, but, usually, it does not produce major microarchitectural disruption. Thus, both a decrease in bone turnover and an increase in hip BMD may be required to decrease hip fracture risk.

thumbnail image

Figure FIG. 1.. Computer generated, three-dimensional surface plot of data from a 1-year RCT in postmenopausal women with osteoporosis reported by Lufkin et al.(9) comparing results in (A) placebo-treatment and (B) treatment with transdermal estrogen. The two plots show the relationship among vertebral fracture rate (VFR), lumbar spine BMD (LS-BMD), and bone formation rate (BFR) assessed by double tetracycline labeling of an iliac biopsy sample. Note that there are two fracture peaks in the placebo-treatment limb, one associated with high bone turnover and one associated with low BMD. By contrast, in the estrogen-treatment limb, the fracture peak associated with high bone turnover has been eliminated whereas the fracture peak associated with the low BMD is still present. Reprinted from Riggs BL, Melton LJ III, O'Fallon WM 1996 Drug therapy for vertebral fractures in osteoporosis: Evidence that decreases in bone turnover and increases in bone mass both determine antifracture efficacy. Bone 18(Suppl):197S-201S, Figure 2, with permission from Elsevier Science.

Download figure to PowerPoint

More recently, Faulkner(12) also addressed the paradox of why small increases in BMD can result in dramatic decreases in vertebral fracture rates. However, he dismisses as unlikely that decreases in bone turnover induced by antiresorptive agents reduce fracture risk independently of their effect on increasing BMD. Instead, he concludes that “bone matters”; that is, the treatment-induced increases in BMD are the cause of the decreased fracture rate induced by antiresorptive agents. He suggests that the paradox can be explained by (1) technical limitations of measuring BMD that result in an underestimation of treatment-induced increases in cancellous bone, (2) differences in the skeletal fragility among study populations, and (3) that the BMD/fracture relationship may not be bidirectional. He also acknowledges the possibility of nondensity-related effects of therapeutics on vision, coordination, muscle strength, and other fall-related factors. However, this possibility is not supported by recent reports that bisphosphonate treatment reduces fractures when BMD is low but not when it is above the threshold for osteoporosis.(18, 19)

Although our 1996 hypothesis and that of Faulkner may not be mutually exclusive, we believe that “bone turnover matters” and that the normalization of high bone turnover is by far the major cause of the discrepant BMD/fracture risk relationship after antiresorptive therapy. Nonetheless, neither we nor Faulkner have explained adequately the first paradox: Why does treatment of osteoporotic patients with raloxifene and calcitonin, which induce only minimal increases in BMD, reduce vertebral fracture risk almost as much as the more potent antiresorptive agents, which induce much larger BMD increases? Thus, we believe that a modification of our 1996 hypothesis is required.

We now suggest that elimination of the disruptive effect of perforative resorption on the microarchitecture of cancellous bone can be achieved by a relatively small effective dosage of an antiresorptive agent but that a larger effective dosage is required to increase BMD further. Thus, both less potent antiresorptive drugs such as raloxifene or calcitonin and more potent ones such as estrogen or bisphosphonates will reduce vertebral fracture risk by an almost comparable degree because it is only necessary to reach a low therapeutic threshold to reduce perforative resorption. However, a greater effective dosage is required to produce larger increases in BMD, which can be achieved using the more potent antiresorptive agents but not the less potent ones. These further increases in BMD confer an additional decrease in fracture risk beyond that achieved by reducing bone turnover, but this decrease is relatively small. Indeed, Sarkar et al.(13) estimated the contribution of increases in BMD to vertebral fracture reduction is 4% for raloxifene therapy whereas Cummings et al.(20) estimated that it is 17% for alendronate therapy. Thus, for substantial BMD-based reductions in vertebral fracture risk to occur, relatively large increases in BMD will be required. Moreover, although population-based studies predict that increasing BMD by 1 SD will reduce fracture risk by one-half; this is unlikely to occur in the treatment of patients with osteoporosis because increases in BMD will not restore the bone microarchitecture that has been disrupted during the process of bone loss. Also, results from many RCTs have shown that even the most potent antiresorptive agents are unlikely to increase BMD by more than 1 SD (10-13%). Thus, the large increases necessary for a substantial BMD-based reduction in fracture risk will require treatment with formation-stimulating agents, such as parathyroid hormone.(21)

Although this modified hypothesis is plausible and seems to agree with observed data better than do conventional concepts, it needs experimental confirmation. This will be difficult to accomplish at present because of the limitations in current technology, but new instrumentation may make testing possible in the near future. Preliminary studies using a prototype instrument for high sensitivity, peripheral three-dimensional quantitative computed tomography, have shown that it is feasible to quantify the degree of microarchitectural disruption of cancellous bone, and the same instrument can make independent measurements of changes in cancellous and cortical BMD.(22) Thus, using this new technology, a head-to-head comparison of a less potent with a more potent antiresorptive agent should soon allow for a rigorous test of this hypothesis.


  1. Top of page
  • 1
    Turner CH 1989 Yield behavior of cancellous bone. J Biomech Eng 111:15.
  • 2
    Marshall D, Johnell O, Wedel H 1996 Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 312:12541259.
  • 3
    Eastell R 1998 Treatment of postmenopausal osteoporosis. N Engl J Med 338:736746.
  • 4
    Riggs BL, Hodgson SF, O'Fallon WM, Chao EYS, Wahner HW, Muhs JM, Cedel SL, Melton LJ III 1990 Effect of fluoride treatment on the fracture rate in postmenopausal women with osteoporosis. N Engl J Med 322:802809.
  • 5
    Lafage M-H, Balena R, Battle MA, Shea M, Seedor JG, Klein H, Hayes WC, Rodan GA 1999 Comparison of alendronate and sodium fluoride effects on cancellous and cortical bone in minipigs. J Clin Invest 95:21272133.
  • 6
    Cosman F, Lindsay R 1999 Selective estrogen receptor modulators: Clinical spectrum. Endocr Rev 20:418434.
  • 7
    Ettinger B, Black DM, Mitlak BH, Knickerbocker RK, Nickelsen T, Genant HK, Christiansen C, Delmas PD, Zanchetta JR, Stakkestad J, Gluer CC, Krueger K, Cohen FJ, Eckert S, Ensrud KE, Avioli LV, Lips P, Cummings SR 1999 Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene results from a 3-year: Randomized clinical trial. JAMA 282:637645.
  • 8
    Chesnut CH, Silverman S, Andriano KA, Genant H, Gimona A, Harris S, Kiel D, LeBoff M, Maricic M, Miller P, Moniz C, Peacock M, Richardson P, Watts N, Baylink D 2000 A randomized trial of nasal spray salmon calcitonin in postmenopausal women with established osteoporosis: The prevent recurrence of osteoporotic fractures study. Am J Med 109:267276.
  • 9
    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 117:19.
  • 10
    Liberman UA, Weiss SR, Broll J, Minne HW, Quan H, Bell NH, Rodrigues-Portales J, Downs RW, Dequeker J, Favus M, Seeman E, Recker RR, Capizzi T, Santor AC, Lombardi A, Shah RV, Hirsch LJ, Karpf DV 1995 Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. N Engl J Med 333:14371443.
  • 11
    Harris ST, Watts NB, Genant HK, McKeever CD, Hangartner T, Keller M, Chesnut CH, Brown J, Eriksen EF, Hoseyni MS, Axelrod DW, Miller PD 1999 Effects of risedronate treatment on vertebral and nonvertebral fractures in women with postmenopausal osteoporosis. JAMA 282:13441352.
  • 12
    Faulkner KG 2000 Bone matters: Are density increases necessary to reduce fracture risk? J Bone Miner Res 15:183187.
  • 13
    Sarkar S, Mitlak BH, Wong M, Stock JL, Black DM, Harper KD 2002 Relationships between bone mineral density and incident vertebral fracture risk with raloxifene therapy. J Bone Miner Res 17:110.
  • 14
    Riggs BL, Melton LJ III, O'Fallon WM 1996 Drug therapy for vertebral fractures in osteoporosis: Evidence that decreases in bone turnover and increases in bone mass both determine antifracture efficacy. Bone 18(Suppl):197S201S.
  • 15
    Parfitt AM, Mathews CHE, Villaneuva AR, Kleerekoper M, Frame B, Rao DS 1983 Relationships between surface, volume, and thickness of iliac trabecular bone in aging and in osteoporosis. J Clin Invest 72:13961409.
  • 16
    Garnero P, Hausherr E, Chapuy MC, Marcelli C, Grandjean H, Muller C, Cormier C, Breart G, Meunier PJ, Delmas PD 1996 Markers of bone resorption predict hip fracture in elderly women: The EPIDOS prospective study. J Bone Miner Res 11:15311538.
  • 17
    Melton LJ III, Khosla S, Atkinson EJ, O'Fallon WM, Riggs BL 1997 Relationship of bone turnover to bone density and fractures. J Bone Miner Res 12:10831091.
  • 18
    Cummings SR, Black DM, Thompson DE, Applegate WB, Barrett-Connor E, Musliner TA, Palermo L, Prineas R, Rubin SM, Scott JC, Vogt T, Wallace R, Yates AJ, LaCroix AZ 1998 Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures. JAMA 280:20772082.
  • 19
    McClung MR, Geusens P, Miller PD, Zippel H, Bensen WG, Roux C, Adami S, Fogelman I, Diamond T, Eastell R, Meunier PJ, Reginster J-Y 2001 Effect of risedronate on the risk of hip fracture in elderly women. N Engl J Med 344:333340.
  • 20
    Cummings SR, Black DM, Person J, Karpf DB, Harris F, Genant HK, LaCroix AZ 1999 How much of the reduction in risk of vertebral fractures by alendronate is explained by increased spine BMD? J Bone Miner Res 14:S159.
  • 21
    Neer RM, Arnaud CD, Zanchetta JR, Prince R, Gaich GA, Reginster J-Y, Hodsman AB, Eriksen EF, Ish-Shalom S, Genant HK, Wang O, Mitlak BH 2001 Effect of parathyroid hormone (1–34) on fractures and bone mineral density in postmenopausal women with osteoporosis. N Engl J Med 344:14341441.
  • 22
    Muller R, Hildebrand T, Hauselmann HJ, Ruegsegger P 1996 In vivo reproducibility of three-dimensional structural properties of noninvasive bone biopsies using 3D-pQCT. J Bone Miner Res 11:17451750.