SEARCH

SEARCH BY CITATION

Aminobisphosphonate treatment has provided clinicians with the means to inhibit bone loss caused by estrogen deficiency, aging and glucocorticoid excess; increase bone mineral density (BMD); and decrease fractures but these potent new agents also have resurrected an old problem, the possibility of “frozen bone.”(1–4) Twenty years ago, Flora et al. examined sequential cortical bone specimens and skeletal radiographs after administration of high-dose bisphosphonates for 1–2 years to adult dogs.(5,6) They found that etidronate caused the accumulation of wide osteoid seams, decreased bone remodeling, and increased the incidence of spontaneous fractures in ribs, dorsal spinal processes of the vertebrae, and the pelvis. Although this skeletal fragility could be blamed on the etidronate-induced defect in mineralization, clodronate also was associated with decreased remodeling and spontaneous fractures in spite of the absence of a long-term effect on osteoid width. So the question was raised, will bones crack and shatter like ice if turnover is halted? Can a significant reduction in remodeling suppress repair, allow microdamage to accumulate, and lead to fatigue-associated fractures? In this issue of the journal, Mashiba et al. offer provocative evidence suggesting that it can.

Other evidence also supports the role of decreased bone remodeling in loss of skeletal integrity (Table 1). Since first described by Albers-Schönberg in 1904, high bone density, defective bone resorption, and fractures with little trauma have distinguished autosomal dominant osteopetrosis (marble bone disease).(7) Although the fracture pattern and healing shows great heterogeneity, 38–88% of patients with type II osteopetrosis, characterized by basilar skull sclerosis, vertebral end plate thickening (“sandwich” vertebrae), and pelvic endobones (radiographic appearance of a bone within a bone), have a history of fractures and 12–62% of patients experience either delayed healing, pseudarthrosis, pathological fractures, or multiple fractures.(8) However, even this evidence is complicated by the recent demonstration of defective osteocalcin production and interleukin-1a (IL-1a)–induced macrophage colony–stimulating factor (M-CSF) release in primary cultures of osteoblastic cells taken from patients with osteopetrosis, possibly the result of aberrant osteoclast-osteoblast interaction.(9) In his brilliant 1957 monograph, Snapper suspected that the transverse character of the “sticks of chalk” fractures in osteopetrosis could not be fully explained by decreased osteoclastic bone resorption (and the resultant accumulation of microdamage) and pointed to abnormal collagen birefringence, the retention of large islands of cartilage within trabeculae and persistent woven bone, as an important cause of the abnormal fragility.(10)

Additional support for the contention that reduced bone remodeling may lead to fractures is supplied by the reports of atypical insufficiency fractures in the femoral neck, sacrum, and pelvis of osteoporotic patients.(11) These fractures are thought to occur in structurally weakened bone that cannot withstand the repetitive mechanical loading of normal activity. Frost suggested that the normal bone multicellular unit was required for repair of this microdamage and that low turnover would retard strain-related repair of microdamage, permit accumulation of microcracks, and thus allow macrodamage to occur.(12) Along these lines, atypical insufficiency fractures are, indeed, associated with low-turnover conditions such as long-term glucocorticoid administration and local bone irradiation and are found more frequently in elderly patients rather than younger patients.(11) However, bone biopsy specimens taken from patients with atypical insufficiency fractures reveal the presence of both high- and low-turnover states, confounding any clear relationship between these fractures and bone remodeling.(11,13)

Evidence also is available suggesting that decreased bone turnover maintains and protects skeletal integrity as shown by the lower incidence of hip fractures in black men and women, a group characterized as exhibiting reduced bone remodeling, and the decreased fractures found in adult patients with type 2 diabetes mellitus, another low-turnover population.(14–16) In addition, profoundly reduced bone turnover occurs in patients with idiopathic hypoparathyroidism and hypothyroidism and yet these patients have increased bone mass and rarely fracture.(17–20) Such patients are said to have reduced exercise tolerance that results in avoidance of fracture situations but patients with untreated hypoparathyroidism often have seizures and decreased ambulatory skills do not usually protect patients with osteoporosis from fracture. The strongest evidence that decreased bone turnover is advantageous is the 47% reduction in vertebral fractures in women treated with alendronate for three years when compared with a group receiving placebo.(21,22) A substantial reduction was seen for clinical as well as morphometric fractures, vertebral and nonvertebral fractures and extended out to four years of therapy.(23,24) Furthermore, long-term bisphosphonate administration has been used in Paget's disease for over 20 years without reports of a late increase in fractures.

Table Table 1.. Decreased Bone Turnover and Increased Risk of Fracture
Evidence forEvidence against
OsteopetrosisBlacks
Atypical insufficiency fracturesType 2 diabetes mellitus
 Hypoparathyroidism and hypothyroidism
 Long-term alendronate treatment

Bone remodeling may be of two types, each with different purpose.(25) One type could be primarily mechanical: deliberate, targeted, dedicated to the maintenance of load-bearing capacity by the prevention or replacement of microdamage, providing a substantial margin of safety, independent of sex steroid levels, and resistant, but not immune, to interference by diseases or drugs. Similar to fracture repair, this type of remodeling would stay functional in almost all situations except osteomalacia. The other type could be primarily metabolic: dedicated to release skeletal calcium stores during late pregnancy, for lactation and in extreme calcium deprivation but otherwise mechanically unnecessary; increased out of proportion to any need for replacement of old bone after loss of sex steroids; and sensitive to hormone replacement therapy, selective estrogen receptor modulators, calcitonin, and aminobisphosphonates. Even if this second type of remodeling was completely shut down, the first type would be adequate to prevent fractures from accumulated microdamage.

Mashiba and colleagues now report that high-dose aminobisphosphonate administration given to young female beagle hounds for 12 months increased the accumulation and length of rib microcracks in proportion to the reduction in activation frequency and decreased bone toughness or the ability to absorb energy although the ultimate force required to fracture the ribs was unchanged.(26) This result was seen with alendronate but not with risedronate. However, the doses administered were five times the clinical amounts used for osteoporosis without any effort to insure dose equivalency. At these clinical doses, preliminary results in patients suggest that alendronate is, indeed, more potent than risedronate with regard to decreasing turnover and increasing bone density, especially at the hip.(1–4,21–24,27) To interpret the findings of Mashiba et al. the reader must ask how do these biomechanical measurements reflect clinical events? What are the most important determinates of bone strength (Table 2)?

Table Table 2.. Determinants of Bone Strength
BMD
Bone size
Cancellous architecture
Bone quality
Osteocyte viability

Although the precise relationship between BMD as determined by dual-energy X-ray absorptiometry (DEXA) and clinically recognized fracture episodes is less clear, in the author's laboratory the power function relationship between in vivo spinal BMD and the force required to cause a vertebral fracture in ex vivo experiments is quite strong (Fig. 1). Other workers also have found that compressive strength in Newtons*1 increases with the square of BMD.(30) Small increases in vertebral BMD result in much greater gains in vertebral strength at least partly because trabecular failure occurs by buckling and bending and the strength of a trabecular strut is proportional to the square of its radius.(31,32) At least 60% of the compression strength is explained by changes in BMD and low BMD is the major contributing factor to skeletal fragility.(33) This strong relationship was still maintained after 2 years of treatment with alendronate in ovariectomized baboons.(30) Moreover, decreased bone turnover with an increased capacity for vertebral stress2 were found after 3 years of treatment in adult beagles using the same 1-mg/kg per day regimen of oral alendronate used by Mashiba et al.(35) The decline in rib toughness3 reported by Mashiba et al. is, therefore, difficult to interpret because so little information is available about the impact of metabolic bone disorders on bone toughness or the clinical significance of measurements done on the ribs versus the vertebra or femoral neck. Furthermore, the strong relationship between biomechanical competence and bone size complicates interpretation of the measurements in growing animals.

thumbnail image

Figure FIG. 1.. The strong positive relationship (x = ayb) between the ultimate load-bearing properties of the murine fifth lumbar vertebrae (in Newtons, as determined by compression between parallel loading platens along the cephalocaudad axis until failure using a material testing machine) and the BMD of the spine (in grams per centimeter squared, as determined in vivo by DEXA(28)). Sixty percent of the compression strength is accounted for by BMD.

Download figure to PowerPoint

thumbnail image

Figure FIG. 2.. The inverted “U”-shaped curve of the relationship between bone turnover and strength.

Download figure to PowerPoint

After BMD, the remaining 40% of the contribution to bone strength is explained by several other characteristics including bone size, cancellous architecture (trabecular separation and number per millimeter) and bone quality. The effect of bone size is revealed by the increase in the risk of femoral neck fracture predicted by the architecturally unsound condition of a long hip axis and the reduced vertebral volume, indicating decreased bone growth during youth, found in postmenopausal women with spinal fractures when compared with those without fractures.(36,37) The role of the cancellous architecture also is easy to understand; an equivalent amount of bone distributed as widely separated, disconnected thick trabeculae is biomechanically less competent than when arranged as more numerous, connected, thin trabeculae.(38,39) This happens because increased connectivity reduces the unsupported length of trabecular struts and strut strength is related inversely to the square of the unsupported length.(31,32) The fractal dimension of cancellous tissue may be another particularly sensitive index of the architecture and predictor of fragility.(40) Additional contributions to bone strength independent of bone size, volume, and architecture involve bone quality.(41) The retained cores of cartilage within trabeculae and woven bone that persist in osteopetrosis represent such qualitative defects.(10) Another well-known example is the marked reduction in bone strength that occurs with long-term fluoride administration in spite of insignificant decreases in ash content or even substantial gains in BMD.(42) More exciting is the recent evidence suggesting that abnormalities in collagen synthesis as revealed by polymorphisms at an Sp1 binding site in the COLIA1 gene, analogous to the collagen defects in osteogenesis imperfecta, predict osteoporotic fracture independently of BMD.(43) However, these features only characterize bone as a flat lifeless matrix, ignoring the three-dimensional microarchitecture and vast communication array of the osteocyte-canaliculi network.(44)

Far from a lifeless mass of mineral and matrix, bone gives shelter to a network of cells. Lining cells, osteoblasts, and osteocytes are connected by cell-to-cell and fluid-filled links in the canalicular system, believed to be a biomechanical sensor array that detects defects in the surrounding bone or changes in external forces and conveys information that results in replacement of substandard materials. Part of the therapeutic efficacy of long-term aminobisphosphonate therapy may be a result of maintenance of osteocyte and osteoblast survival with preservation of the biomechanical sensor system. Recent evidence shows that bisphosphonates promote osteoblast and osteocyte survival by preventing apoptosis. The antiapoptotic effect is associated with a rapid increase in phosphorylation of extracellular signal–regulated kinases (ERKs) and is blocked by specific inhibitors of ERK activation.(45) Therefore, aminobisphosphonate therapy may preserve not only the acellular but also the cellular portions of bone.

At some very high dose, not yet attainable in clinical practice, bisphosphonates may shut down remodeling, but this would probably be reversible. All evidence indicates that these agents can be used for extended periods without fear of fractures in humans. However, the potential of bringing remodeling to a standstill indicates that monitoring of long-term bisphosphonate treatment should be continued.

The relationship between bone turnover and strength seems to have an inverted “U”-shaped curve (Fig. 2). Accelerated turnover increases fragility because of osteoid accumulation, decreased secondary mineralization, and increased erosion bays that represent temporarily weakened, focal cancellous lesions especially dangerous in load-bearing, vertically oriented, vertebral trabeculae.(46–48) Absent turnover also is bad because some minimal amount of remodeling is necessary to repair fatigue microdamage, replace old or dead osteocytes, and restore bone hydration, but existing evidence indicates that instead of increasing skeletal fragility, there is a considerably wide safety margin of decreased bone turnover associated with increased bone strength. If bone quality is unimpaired, the most important determinant of bone strength is BMD. True bone strength is the resistance to fracture, the clinical endpoint of all osteoporosis research.

Acknowledgements

  1. Top of page
  2. Acknowledgements
  3. REFERENCES

I am grateful to A. Michael Parfitt for his advice on the manuscript and to William R. Hogue, University of Arkansas for Medical Sciences Department of Orthopedics, for help with the biomechanical measurements. This work was supported by the National Institutes of Health (PO1AG13918 and RO1AR46191) and a Research Enhancement Award Program (REAP) grant and VA Merit Review grant from the Veterans Administration.

  • 1

    * A Newton is a unit of force (F = k m a, where k is a constant, m is mass, and a is acceleration), so that 1 N is the force that will give a mass of 1 kg an acceleration of 1 m/s2. For example, a falling 1-kg weight accelerates at 9.8 m/s2 and so the force pulling down on it is 9.8 N. A falling 10-kg weight is pulled by 98 N. A one-quarter pound block of butter on your hand represents a downward force or load of 1 N.(29)

  • 2

    Stress is defined as force per unit area and is expressed in Pascals (1 Pascal = 1 N/m2). In bone, stress is measured as the ultimate force required to fracture divided by the cross-sectional moment of inertia, a measure of the geometric distribution of material around a given axis.

  • 3

    Toughness is the ability to absorb energy or sustain deformation without breaking and is defined as the area under the curve of load or force versus displacement, also known as the work to failure, divided by the cross-sectional moment of inertia.(34) Bone stiffness or rigidity is defined as the slope of the linear portion of the load-displacement curve. Strain is the relative deformity induced by a load or the area under the load-displacement curve (1000 microstrain [μϵ] equals a 0.1% deformation).

REFERENCES

  1. Top of page
  2. Acknowledgements
  3. REFERENCES
  • 1
    Liberman UA, Weiss SR, Broll J, Minne HW, Quan H, Bell NH, Rodriguez-Portales J, Downs RW, Dequeker J, Favus M, Seeman E, Recker RR, Capizzi T, Santora AC, Lombardi A, Shah RV, Hirsch LJ, Karpf DB 1995 Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. N Engl J Med 333: 14371443.
  • 2
    McClung M, Clemmesen B, Daifotis A, Gilchrist N, Eisman J, Weinstein RS, Fuleihan GEH, Reda C, Yates AJ, Ravn P 1998 Alendronate prevents postmenopausal bone loss in women without osteoporosis. Ann Intern Med 128: 253261.
  • 3
    Bone HG, Downs RW, Tucci JR, Harris ST, Weinstein RS, Licata AA, Kimmel DB, Gertz BJ, Hale E, Polvino WJ 1997 Dose-Response Relationships for Alendronate Treatment in Osteoporotic Elderly Women. J Clin Endocrinol Metab 82: 265274.
  • 4
    Saag KG, Emkey R, Schnitzer TJ, Brown JP, Hawkins F, Goemaere S, Thamsborg G, Liberman UA, Delmas PD, Malice MP, Czachur M, Daifotis AG 1998 Alendronate for the prevention and treatment of glucocorticoid-induced osteoporosis. N Engl J Med 339: 292299.
  • 5
    Flora L, Hassing GS, Parfitt AM, Villanueva AR 1980 Comparative skeletal effects of two diphosphonates in dogs. Metab Bone Dis Rel Res 2S: 389407.
  • 6
    Flora L, Hassing GS, Cloyd CG, Bevan JA, Parfitt AM, Villanueva AR 1981 The long-term effects of EHDP in dogs. Metab Bone Dis Rel Res 4 and 5: 289300.
  • 7
    Albers-Schönberg H 1904 Röntgenbilder einer seltenen Knochenerkrankung. Muenchener Med Wschr 51: 365.
  • 8
    Bollerslev J 1989 Autosomal dominant osteopetrosis: Bone metabolism and epidemiological, clinical, and hormonal aspects. Endocr Rev 10: 4567.
  • 9
    Lajeunesse L, Busque L, Ménard P, Brunette MG, Bonny Y 1996 Demonstration of an osteoblast defect in two cases of human malignant osteopetrosis: Correction of the phenotype after bone marrow transplantation. J Clin Invest 98: 18351842.
  • 10
    Snapper I 1957 Chapter 24: Osteopetrosis, Albers-Schönberg disease, marble bone disease. In: SnapperI (ed.) Bone Disease in Medical Practice. Grune & Stratton, New York, NY, U.S.A., pp. 175179.
  • 11
    McKenna MJ, Kleerekoper M, Ellis BI, Rao DS, Parfitt AM, Frame B 1987 Atypical insufficiency fractures confused with Looser Zones of osteomalacia. Bone 8: 7178.
  • 12
    Frost HM 1963 Introduction to Biomechanics. Charles C. Thomas, Springfield, IL, U.S.A.
  • 13
    Perry HM, Weinstein RS, Teitelbaum SL, Avioli LV, Fallon MD 1982 Pseudofractures in the absence of osteomalacia. Skeletal Radiol 8: 1719.
  • 14
    Weinstein RS, Bell NH 1988 Diminished rate of bone formation in normal black adults. N Engl J Med 319: 1689701.
  • 15
    Heath H, Melton LJ, Chu C-P 1980 Diabetes mellitus and the risk of skeletal fracture. N Engl J Med 303: 567570.
  • 16
    Van Daele PLA, Stolk RP, Burger H, Algra D, Grobbee DE, Hofman A, Birkenhager JC, Pols HAP 1995 Bone density in non-insulin-dependent diabetes mellitus: The Rotterdam study. Ann Intern Med 122: 409414.
  • 17
    Eriksen EF, Mosekilde L, Melsen F 1986 Kinetics of trabecular bone resorption and formation in hypothyroidism: Evidence for a positive balance per remodeling cycle. Bone 7: 101108.
  • 18
    Albright F, Reifenstein EC 1948 Chapter 2: Clinical hypoparathyroidism. In: AlbrightF, ReifensteinEC (eds.) The Parathyroid GlsMetabolic Bone Disease. Williams & Wilkins, Baltimore, MD, U.S.A., pp. 2740.
  • 19
    Bronsky D, Kushner DS, Dubin A, Snapper I 1958 Idiopathic hypoparathyroidism and pseudohypoparathyroidism: Case reports and review of the literature. Medicine (Baltimore) 37: 317352.
  • 20
    Parfitt AM 1979 Surgical, idiopathic, other varieties of parathyroid hormone-deficient hypoparathyroidism. In: DeOrootU, CahillOFJr, OdellWD (eds.) Endocrinology. Grune & Stratton, New York, NY, U.S.A., pp. 755768.
  • 21
    Black DM, Cummings SR, Karpf DB, Cauley JA, Thompson DE, Nevitt MC, Bauer DC, Genant HK, Haskell WL, Marcus R, Ott SM, Torner JC, Quandt SA, Reiss TF, Ensrud KE 1996 Randomized trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet 348: 15351541.
  • 22
    Ensrud KE, Black DM, Palermo L, Bauer DC, Barrett-Conner E, Quandt SA, Thompson DE, Karpf DB 1997 Treatment with alendronate prevents fractures in women at highest risk. Arch Intern Med 157: 26172624.
  • 23
    Karpf DB, Shapiro DR, Seeman E, Ensrud KE, Johnston CC, Adami S, Harris ST, Santora AC, Hirsch LJ, Oppenheimer L, Thompson D 1997 Prevention of nonvertebral fractures by alendronate: A meta-analysis. JAMA 277: 11591164.
  • 24
    Cummings SR, Black DM, Thompson DE, Applegate WB, Barrett-Conner 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: Results from the fracture intervention trial. JAMA 280: 20772082.
  • 25
    Parfitt AM 1996 Chapter 12: Skeletal heterogeneitythe purposes of bone remodeling. In: MarcusR, FeldmanD, KelseyJ (eds.) Osteoporosis. Academic Press, San Diego, CA, U.S.A., pp. 315329.
  • 26
    Mashiba T, Hirano T, Turner CH, Forwood MR, Johnston CC, Burr DB Suppressed bone turnover by bisphosphonates can increase microdamage accumulation and reduce biomechanical properties in dog rib. J Bone Miner Res (in press)
  • 27
    Mortensen L, Charles P, Bekker PJ, Digennaro J, Johnston CC 1998 Risedronate increases bone mass in an early postmenopausal population: Two years of treatment plus one year of follow-up. J Clin Endocrinol Metab 83: 396402.
  • 28
    Jilka RL, Weinstein RS, Takahashi K, Parfitt AM, Manolagas SC 1996 Linkage of decreased bone mass with impaired osteoblastogenesis in a murine model of accelerated senescence. J Clin Invest 97: 17321740.
  • 29
    Rogers EM 1960 Part One: Matter, motion, force. In: Rogers, EM (ed.) Physics for the Inquiring Mind: The Methods, Nature, Philosophy of Physical Science. Princeton University Press, Princeton, NJ, U.S.A., pp. 3190.
  • 30
    Balena R, Toolan BC, Shea M, Markatos A, Myers ER, Lee SC, Opas EE, Seedor JG, Klein H, Frankenfield D, Quartuccio H, Fioravanti C, Clair J, Brown E, Hayes WC, Rodan GA 1993 The effects of 2-year treatment with the aminobisphosphonate alendronate on bone metabolism, bone histomorphometry and bone strength in ovariectomized nonhuman primates. J Clin Invest 92: 25772586.
  • 31
    Bell GH, Dunbar O, Beck JS, Gibb A 1967 Variations in strength of vertebrae with age and their relation to osteoporosis. Calcif Tissue Res 1: 7586.
  • 32
    Carter DR, Hayes WC 1977 The compressive behavior of bone as a two-phase porous structure. J Bone Joint Surg Am 59: 954962.
  • 33
    Keen RW 1999 Effects of lifestyle interventions on bone health. Lancet 354: 19231924.
  • 34
    Turner CH, Burr DB 1993 Basic biomechanical measurements of bone: A tutorial. Bone 14: 595608.
  • 35
    Balena R, Markatos A, Guy J, Shea M, Seedor JG, Hayes WC, Peter CP, Rodan GA 1993 Effects of three years treatment with alendronate in adult beagles. J Bone Miner Res 8: S270 (abstract).
  • 36
    Faulkner KG, McClung M, Cummings SR 1994 Automated evaluation of hip axis length for predicting hip fracture. J Bone Miner Res 9: 10651070.
  • 37
    Duan Y, Parfitt AM, Seeman E 1999 Vertebral bone mass, size and volumetric density in women with spinal fractures. J Bone Miner Res 14: 17961802.
  • 38
    Kleerekoper M, Villanueva AR, Stanciu J, Rao DS, Parfitt AM 1985 The role of three-dimensional trabecular microstructure in the pathogenesis of vertebral compression fractures. Calcif Tissue Int 37: 594597.
  • 39
    Weinstein RS, Hutson MS 1987 Decreased trabecular width and increased trabecular spacing contribute to bone loss with aging. Bone 8: 137142.
  • 40
    Weinstein RS, Majumdar S 1994 Fractal geometry and vertebral compression fractures. J Bone Miner Res 9: 17971802.
  • 41
    Weinstein RS, Sappington LJ 1982 Qualitative bone defect in uremic osteosclerosis. Metabolism 31: 805811.
  • 42
    Sogaard CH, Mosekilde Li, Richards A, Mosekilde Le 1994 Marked decrease in trabecular bone quality after five years of sodium fluoride therapy—assessed by biomechanical testing of iliac crest bone biopsies in osteoporotic patients. Bone 15: 393399.
  • 43
    Langdahl BT, Ralston SH, Grant SFA, Eriksen EF 1998 An Sp1 binding site polymorphism in the COLIA1 gene predicts osteoporotic fractures in both men and women. J Bone Miner Res 13: 13841389.
  • 44
    Weinstein RS, Manolagas SC 2000 Apoptosis and Osteoporosis. Am J Med 108: 153164.
  • 45
    Plotkin LI, Weinstein RS, Parfitt AM, Roberson PK, Manolagas SC, Bellido T 1999 Prevention of osteocyte and osteoblast apoptosis by bisphosphonates and calcitonin. J Clin Invest 104: 13631374.
  • 46
    Parfitt AM 1993 Pathophysiology of bone fragility. Fourth International Symposium on OsteoporosisConsensus Development Conference, Helstrykkeriet Aalborg ApS, Aalborg, Denmark, pp. 164166.
  • 47
    Riggs BL, Melton LJ, O'Fallon WM 1993 Toward optimal therapy of established osteoporosis: evidence that antiresorptiveformation-stimulating regimens decrease vertebral fracture rate by independent mechanisms. Fourth International Symposium on OsteoporosisConsensus Development Conference, Helstrykkeriet Aalborg ApS, Aalborg, Denmark, pp. 1315.
  • 48
    Lafage M-H, Balena R, Battle MA, Shea M, Seedor JG, Klein H, Hayes WC, Rodan GA 1995 Comparison of alendronate and sodium fluoride effects on cancellous and cortical bone in minipigs: A one-year study. J Clin Invest 95: 21272133.