1. Top of page
  2. Abstract
  8. References

Bone strength and “mass” normally adapt to the largest voluntary loads on bones. The loads come from muscles, not body weight. Bone modeling can increase bone strength and “mass,” bone remodeling can conserve or reduce them, and each can turn ON and OFF in response to its own threshold range of bone strains. During growth, the loads on bones from body weight and muscle forces increase, and modeling correspondingly increases bone strength and “mass.” In young adults those loads usually plateau, so bone strength can “catch up” and modeling can turn OFF. Meanwhile remodeling keeps existing bone. After about 30 years of age, muscle strength usually decreases. In aging adults this would put bones that had adapted to stronger young-adult muscles into partial disuse and make remodeling begin to reduce their strength and “mass,” as disuse regularly does in experimental situations in other mammals, both growing and adult. Those changes associate strongly with the size of the bone strains caused by the loads on bone. While nonmechanical effects associated with aging should contribute to that age-related bone loss too, a new skeletal paradigm suggests the above mechanical influences would dominate control of the process in time and anatomical space.


  1. Top of page
  2. Abstract
  8. References

MANY PAST EXPLANATIONS for the slow loss of bone in most aging humans invoked nonmechanical factors, including in part gradual losses of bone-cell responsiveness to hormonal and dietary influences and decreasing supplies of the precursor cells that create osteoblasts and osteoclasts (Table 1).1–3 A new skeletal-biologic paradigm adds a vital-biomechanical explanation to those ideas.4,5 This text briefly summarizes that explanation and some old and new ideas and evidence it depends on. The text appends a Glossary and an explanatory Note.

Thumbnail image of


  1. Top of page
  2. Abstract
  8. References

Bone strength and “mass”: Physical determinants and relationships

Bone's “material properties” partly determine its strength.6 They include its stiffness, ultimate strength, yield point, and true density. However, they change little with age, gender, species, and most diseases compared with the “mass” and architectural contributions described next. The amount of bone in a cross-section (the “mass” contribution) affects the strength of a whole bone. Usually the more bone, the stronger it is. A bone's outside diameter and shape and the distribution of its cortical and trabecular bone (the architectural contribution) also affect its strength. For example, doubling a bone's diameter without changing the amount of bone in its cross-section increases its strength in bending eight times.6,7 While stronger bones usually have more bone “mass” too, in some fracture mechanisms the architectural contribution to bone strength can be more important than the mass contribution.8 With suitable software, peripheral quantitative computed tomography can evaluate both contributions better than current single or dual beam absorptiometric methods,7,9–13 so it may see increasing use for that purpose.

The usual aging pattern:

Normally human bone “mass” increases during growth, plateaus in young adult life, and after about 35 years of age, it begins to decrease. At 70 years of age, less than 70% of the young-adult bone “mass” can remain.14,15 Yet the minor fraction of aging people who continue to do hard physical work (farmers, loggers, construction steel workers, weight lifting enthusiasts) keep their bone better than most others and much better than aging adults who become inactive physically because of medical and allied problems like those listed in Table 2.

Thumbnail image of

Biologic determinants of bone mass

Global bone modeling by drifts (not osteoblasts alone) provides the chief mechanism for increasing bone strength and “mass” (Fig. 1).4,6,16 Where bone strains exceed a threshold range, this modeling usually turns ON and begins to increase bone strength and “mass”. Where strains stay below this “modeling threshold” or minimum effective strain range (MESm; centered near 1000 μstrain), modeling usually turns OFF. 11 Modeling does not decrease bone strength and “mass,” and modeling of cortical bone becomes inefficient in adults.17 “Global” means averaged over a whole bone or skeleton. Global bone remodeling by BMUs (basic multicellular units) provides nature's chief mechanism for turning bone over, for removing mechanically unneeded bone, and for repairing microdamage (microscopic fatigue damage) (Fig. 2) (it has other functions too). Where bone strains exceed a smaller threshold range, bone remodeling functions in its “conservation mode” (Fig. 2), in which the amounts of bone resorbed and made by completed BMUs tend to equalize. Where strains stay below this “remodeling threshold” or MESr (centered near 50–100 μstrain) its conservation mode changes to its “disuse mode” in which completed BMUs make less bone than they resorb (Fig. 2). This makes them begin to remove bone permanently, usually where it touches marrow. A typical adult “disuse-pattern osteopenia” results, characterized by normal bone lengths and outside diameters but with widened marrow cavities that thin the cortices, and by reduced amounts of spongiosa. Partial and/or gradual onset disuse causes similar effects on bone strength and “mass,” and a similar osteopenia, but so slowly that they can take several years to become obvious. In addition to causing net bone loss, acute partial disuse usually increases bone turnover.

thumbnail image

Figure FIG. 1. Bone modeling drifts. (A) Diagrams an infant's long bone with its original size and shape in solid line. To keep this shape as it grows in length and diameter, its surfaces must move in tissue space as the dashed lines suggest. Formation drifts make and control osteoblasts to build some surfaces up, while separate resorption drifts make and control osteoclasts to remove material from other surfaces. (B) A different drift pattern can correct the fracture malunion in a child, shown in solid line. The cross-section view to the right shows the cortical-endosteal as well as the periosteal drifts that do that. (C) Shows how the drifts in (B) would move the whole segment to the right in tissue space (reproduced by permission: Frost HM 1987 Osteogenesis imperfecta. The setpoint proposal. Clin Orthop 216:280–297).

Download figure to PowerPoint

thumbnail image

Figure FIG. 2. Bone remodeling BMUs. (Top row) An activation event on a bone surface at (A) causes a packet of bone resorption by osteoclasts at (B), and then replacement of the resorbed bone by osteoblasts at (C) on the right. The BMU makes and controls the ‘clasts and’ blasts that do this. (Second row) Idealize those events to emphasize the amounts of bone resorbed (E) and formed (F) by completed BMUs. (Third row) In these “BMU graphs” (after the author), (G) on the left shows a small excess of formation over resorption as on periosteal surfaces. (H) shows equalized resorption and formation as on haversian surfaces (this is conservation-mode remodeling). (I) on the right shows a net deficit of formation, as on cortical-endosteal and trabecular surfaces (this is disuse-mode remodeling). (Bottom row) These “stair graphs” (after P.J. Meunier) show the effects on the local bone balance and “mass” of a series of BMUs of the kind immediately above (reproduced by permission: Frost HM 1987 Osteogenesis imperfecta. The setpoint proposal. Clin Orthop Rel Res 216:280–297).

Download figure to PowerPoint

The main purpose of those activities should lie in making bones strong enough so the usual voluntary loads on them do not break them or make them hurt. In that process, bone “mass” provides a useful if not always reliable surrogate of bone strength.7,11 The size of the bone strains and their locations on bone surfaces should provide at least one means the biologic mechanisms can use to monitor and evaluate a bone's strength relative to the mechanical loads on it.7,18 The above arrangement would adjust an intact bone's strength to the usual largest voluntary loads on it to keep typical peak strains from exceeding the above modeling threshold. This would mean big bones became big to keep big voluntary loads from causing strains above that threshold.

The above features could explain why, when physical activities change from normal to disuse, modeling usually turns OFF and disuse-mode remodeling turns ON, eventually to cause an osteopenia, in children and adults. The features could explain why, when normal physical activities change to hypervigorous ones (as in weight lifting17), modeling turns ON in children, and conservation-mode remodeling turns ON in both children and adults. This further increases bone mass in children, but it mainly conserves existing bone in aging adults. These facts are clear at the clinical and organ levels.15,19 W.S.S. Jee's group, among others, has shown how the tissue-level responses of modeling and remodeling to mechanical influences cause these observations.20–24 Further research needs to study the cell- and molecular-biologic mechanisms these responses depend on.

Some roles of muscle physiology

Trauma excepted, the largest voluntary loads on bone come from muscle contractions, not body weight. To move us around on Earth, muscles must fight the resistance of the body's weight multiplied by the bad lever arms most muscles work against.6,25 Therefore, it takes over 2 lb of muscle force on bones to move each 1 lb of body weight around as we work and play. Also, as a 7 lb birth weight increases over 25 times to become a 200 lb young adult, the muscle forces exerted to work and play against that increased resistance increase correspondingly. Such facts help to explain why the voluntary total load on a football player's femur during a game can briefly exceed five times body weight.26,27 During growth, modeling adapts bone strength to such loads to keep them from breaking bones. Since it takes stronger muscles to move heavier bodies around, bone “mass” should correlate well with body weight in normally active individuals, as it does,14 but poorly in bed-ridden individuals with normal or increased body weight.

Therefore, voluntary muscle forces should dominate a bone's postnatal structural adaptations to its mechanical usage, modified somewhat by body weight and one's voluntary physical activities. It seems clear that they do, as D'Arcy Thompson suggested long ago.28 This should make things that affect muscle strength important in understanding bone strength and “mass.”4,29–31 In part those things include the growth hormone, androgens, calcium, and vitamin D that many reviews of bone physiology discuss, but mainly in the context of their effects on bone cells instead of on muscle and the vital biomechanics.3,14

Usually muscle strength increases during growth, it plateaus in young adults, and after 30–40 years of age it begins a decline that continues for the rest of life, so at 80 years of age less than half of one's young-adult muscle strength may remain.2,32–37 The age-related changes in bone “mass” usually follow these age-related changes in muscle strength.

Clinical and experimental observations show that muscle strength can usually change more rapidly than the sluggish modeling and remodeling processes can change bone strength and “mass.” Therefore, when muscle strength keeps changing in the same direction (i.e., when it keeps increasing as in childhood, or decreasing as in aging adults) the corresponding adaptations in bone strength and “mass” should tend to lag behind, and could catch up to and fit the mechanical need only after muscle strength had plateaued for some period of time.

If so, comparisons of reliable bone strength indices (BSIs) to the measured strength of muscles loading the bone in question might show a modest deficit in bone strength during growth, and an excess of it in aging adults compared with young adults between 20–30 years of age. This idea could be tested. One can easily measure muscle strength in humans, preferably as the maximum torque generated about joints like the knee, hip, elbow, and fingers.12,13,33 Noninvasive peripheral quantitative computed tomography can provide BSIs that account quite well for both the “mass” and architectural contributions to bone strength.7,9–12,29 The few such comparisons done so far were very informative.

Combined modeling, remodeling, and muscle strength effects

In Fig. 3, the horizontal line at the bottom of the graph suggests typical peak bone strains from zero on the left, to the fracture strain on the right (Fx ≈ 25,000 microstrain), plus the locations of the remodeling, modeling, and microdamage threshold ranges (MESr, MESm, MESp). (This text does not discuss microdamage but its threshold range in bone centers near 3000 μstrain.) The horizontal axis represents no net gains or losses of bone strength or mass. The lower dotted line curve suggests how remodeling would remove and weaken bone where strains fall to or below the MESr range and stay there, but otherwise would begin to keep existing bone. The upper dashed line curve suggests how modeling would increase bone strength and mass where strains enter or exceed the MESm range. The dashed outlines suggest the combined modeling-remodeling effects (D.H. Carter originally suggested such a curve). At and beyond the MESp range, woven bone formation drifts usually replace lamellar bone formation drifts. Fx = the fracture strain near 25,000 μstrain. At the top, DW = disuse window; AW = adapted window or “comfort zone” as in normally adapted adults; MOW = mild overload window, as in growing mammals and children; and POW = pathologic overload window.4

thumbnail image

Figure FIG. 3. Combined modeling, remodeling, and muscle effects. In the nearly flat “comfort zone” or adapted window between the MESr and MESm threshold ranges, bone strength and “mass” change little as typical strains change. In children increasing body weight and muscle loads on bone should up-shift bone strains towards the MESm (to the right) and turn modeling ON. In most aging adults, decreasing muscle strength should down-shift the strains toward the MESr (to the left), to turn disuse-mode remodeling ON and cause a slow loss of bone (adapted from: Frost HM 1994 Perspectives: A vital biomechanical model of synovial joint design. Anat Rec 240:1–18).

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  8. References

Usually our muscle strength increases during growth, plateaus in young adult life, and begins to decline after about 30 years of age. Since muscles apply the largest loads on bones, and since bones normally adapt their strength and “mass” to the largest loads on them, bone strength and “mass” should also increase during growth, plateau in young adults, and decrease after about 30 years of age, as they do.

Since bone strength and “mass” plateau in most young adults, their peak bone strains should downshift from the modeling threshold region (MESm) where they usually occur during growth, into the comfort zone in Fig. 3. There modeling would turn OFF but conservation-mode remodeling would continue. This could explain the observed plateau in bone strength and “mass” in young adults.

In aging adults, decreasing muscle strength would decrease the loads on bones that had adapted to stronger young-adult muscles. That would put such bones in partial and gradual-onset disuse. This should further downshift their strains into the remodeling threshold region (MESr) in Fig. 3, to make disuse-mode remodeling cause a slow loss of bone next to marrow. Relative to young-adult bone strength and “mass,” a typical disuse-pattern osteopenia would develop.

This could explain why, in most aging adults, disuse-mode remodeling does turn ON and modeling does remain OFF to cause gradual losses of bone strength and “mass”. As long as muscle strength keeps decreasing, those losses should continue.


  1. Top of page
  2. Abstract
  8. References

On the effects of nonmechanical factors

The above explanation would supplement others based on changes in the nonmechanical factors that could also affect bone strength and “mass” in aging adults. Many such factors exist (Table 1).1,38 However, the paradigm would propose that: (1) the mechanical usage responses need nonmechanical factors in order to work, as cars need wheels, motors and fuel; (2) but mechanical usage, and especially strain, guide those responses in time and anatomical space, as steering, brakes, accelerators, and ignition switches do in cars; and (3) most (but not all) nonmechanical influences could only help or hinder the mechanical usage responses and could not replace them nor control them in time and space (otherwise they could normalize bone architecture, strength, and “mass” in paralyzed limbs). While that opens the issue for discussion, in support of it experienced clinicians know that aging adults who keep their young-adult muscle strength better than others, also keep their young-adult bone strength and “mass” better than others. Such adults do hard physical work like logging, farming, construction, steel working, and weight lifting.15,19,39,40

On prevention

The above material suggests several general approaches to preventing our age-related bone loss. They would include, in part: (A) use medications to depress disuse-mode remodeling, since it (not osteoclasts alone) is the specific biologic mechanism that removes the bone (bisphosphonates?, estrogen?); (B) engage in appropriate physical exercise (weight lifting or equivalents) to maintain young-adult muscle strength during subsequent aging; (C) use medications that maintain muscle strength even when an individual's physical activities decrease (androgens, growth hormone, or chemical derivatives of them?); (D) or, find medications that potentiate or duplicate bone's normal responses to hypervigorous mechanical usage (none yet known, but none specifically sought either). Discussion of the pros and cons of those and other possible approaches seems best deferred to another time and place.

On estrogen effects

The new paradigm suggests explanations for some features of the bone loss that happens in women during and after menopause.14 That process begins with increases in bone turnover by BMU-based remodeling and in net bone loss. After 5 or so years, that removes about 15% of the former bone “mass.” Then bone turnover and net bone losses decrease to lower levels and tend to remain there afterwards.41 The bone that is lost comes mainly from bone touching marrow (trabeculae and endocortical bone), not from subperiosteal bone, so the final osteopenia is characterized by normal bone lengths and outside diameters, but widened marrow cavities that thin the cortices, and reduced amounts of spongiosa.14 Those tissue-dynamic and anatomical bone loss patterns precisely copy the acute and chronic disuse patterns,4 which suggests that estrogen and disuse may both act on the same controlling mechanism(s).

It has been suggested that estrogen could lower the modeling and remodeling thresholds.6,42,43 That would tend to increase bone “mass” and conserve it, by making smaller strains than before turn modeling and conservation-mode remodeling ON. If so, removing estrogen would raise those thresholds. That would make bone “sense” partial disuse and turn modeling OFF, and make disuse-mode remodeling begin to remove the “excess” bone until the remainder fits the new thresholds. Then further net losses would reduce to age-comparable norms, so when compared with the premenopausal state an adult disuse-pattern osteopenia would be the end result. It was also suggested that estrogen's bone effects had the purpose of storing extra calcium as bone to meet the needs of lactation after a pregnancy. When menopause made pregnancy no longer possible, that extra bone, which according to this idea would not be needed for mechanical reasons, is removed.4 Further discussion of those ideas and of the roles of estrogen effects on osteoclasts, osteoblasts and their precursor cells, and of their purposes, also seems best deferred to another time and place.

  • 1

    On in vivo strains. Before in vivo strain studies were available, the author inferred from clinical-anatomical-pathologic evidence: (A) modeling drifts add bone mass and strength; (B) bone strains control the drifts; (C) the drifts respond to some average of those strains instead of to rare or single ones; (D) a few large strains have more effect than small ones no matter how frequent; (E) changing instead of constant strain controls drifts; (F) some minimum effective strain (an MES) switches drifts ON; otherwise they would stay OFF; (G) modeling is rate limited and cannot exceed a maximum speed. While controversial at the time,44 subsequent studies in many animals including humans verified all these ideas and current work concerns their values.18,31,45–49 Such studies show that peak cortical bone strains from voluntary efforts can reach 2000–4000 μstrain in most rapidly growing subjects, but in most adults they range between about 800–1300 μstrain.

  • 1

    Taken from Ref. 4.


  1. Top of page
  2. Abstract
  8. References

1 In the current literature, some terms have variable or even vague meanings. The meanings of some such terms in this text follow.

bone “mass”: the amount of bone tissue, often estimated by absorptiometry, and preferably viewed as a volume minus the marrow cavity. It does not mean gravimetric mass in the sense used in physics. It can provide a sometimes unreliable index of bone strength because it does not account for the architectural contribution to bone strength. When in quotes in this text, “mass” means its use in absorptiometry.

BMU: basic multicellular unit of bone remodeling. In approximately 4 months, and in a biologically coupled activation [RIGHTWARDS ARROW] resorption [RIGHTWARDS ARROW] formation or ARF sequence, it turns over approximately 0.05 mm3 of bone (Fig. 2). When it makes less bone that it resorbs, this tends to remove bone permanently, usually next to marrow. Adult humans may create about 3 million new BMUs annually, and about a million may function at any moment in the whole skeleton.(4)

disuse: the meaning of disuse may seem quite clear to us, but bones cannot read minds. They need another way to define it. When typical peak bone strains down-shift into the remodeling threshold region in Fig. 3, from a bone's point of view that would represent disuse and signal its existence, no matter how small or big the bone. In such situations disuse-mode remodeling usually turns ON to reduce local bone strength and “mass.”

drift: see “modeling” below.

osteopenia: less bone that usual for most healthy people of the same age, height, weight, gender, and race. Its definition still involves problems and different opinions this text does not discuss. Relative to their young-adult bone “mass,” most aged adults have an osteopenia, and it is the subject of this article.

MESm: minimum effective strain range (or equivalent stimulus) for switching mechanically controlled bone modeling drifts ON. An operational concept, the center of this genetically determined “modeling threshold” range probably lies near 1000 μstrain in most adults, which one could view as its set point (it corresponds to approximately 20 MPa). Its value currently causes some debate, but its existence, suggested by the author in 1964 (footnote 1F), no longer does. Nonmechanical agents, disease, age, race, and species might modify its value.

MESr: the minimum effective strain range (or equivalent stimulus) for the mechanical control of BMU-based remodeling. Above it BMU creations begin to decrease, and completed BMUs tend to make and resorb equal amounts of bone, which defines conservation-mode remodeling. Where strains stay below the MESr, BMU creations increase and completed BMUs make less bone than they resorb. This disuse-mode remodeling increases local bone loss (usually next to marrow), decreases local bone strength and causes an osteopenia. Another operational concept, the center of this genetically determined but less studied threshold range may lie near 50–100 μstrain, which one could view as its set point (it corresponds to ≈ 1–2 MPa). Nonmechanical agents, disease, age, race and species might modify its value.

modeling: the biologic processes that produce functionally purposeful sizes and shapes to skeletal organs. Mostly independent resorption and formation modeling drifts do it in bones. A resorption drift progressively removes bone from a surface without any associated formation. A formation drift progressively adds bone to another bone surface without any associated resorption. Modeling drifts mainly determine outside bone diameter, cortical thickness and the upper limit of bone strength and “mass.” Remodeling determines the lower limits (Fig. 1).

osteopenia: less bone that normal, and therefore weaker and more likely than normal to fracture from an injury. It need not always stem from an intrinsic bone disease.

remodeling: turnover of bone in small packets by BMUs. Pre- 1964 literature did not distinguish between modeling and remodeling and lumped them together as remodeling. Some authors still do that, which can be confusing. However, while drifts and BMUs create and use what seem to be the same kinds of osteoblasts and osteoclasts to do their work, in different parts of the same bone at the same time ‘blasts and’ clasts in drifts and BMUs can respond differently and even oppositely to the same influence.(5,6,30,21,23,24) In remodeling's “disuse mode,” BMU creations increase and completed BMUs make less bone than they resorb. In its “conservation mode” BMU creations usually decrease and resorption and formation in completed ones tend to equalize. A pearl: Where bone strength and “mass” increase, modeling did it, not osteoblasts alone; where they decrease, disuse-mode remodeling did it, not osteoclasts alone.

resorption: different meanings of this term in the literature cause some confusion. Many authors use it to mean net bone loss, and in that sense discuss “antiresorption agents.” Others use it to mean bone resorption by osteoclasts, and refer to net losses of bone as such and separately. It means resorption by osteoclasts in this text. In that sense few true antiresorption agents exist and they do not include estrogen or presently known and studied bisphosphonates, which instead are antiremodeling agents that decrease BMU creations. Due to the ARF sequence, this reduces global resorption first, and then formation, and both about equally.

strain: the deformation or change in dimensions and/or shape caused by a load on any structure or structural material. Special gages can measure bone strains in the laboratory and in vivo. Loads always cause strains, even if very small ones. Biomechanicians often express strain in microstrain units, where 1000 μstrain in compression would shorten a bone by 0.1% of its original length, 10,000 μstrain would shorten it by 1% of that length, and 100,000 μstrain would shorten it by 10% of that length (and break it). Above approximately 3000 μstrain bone's stress-strain curve is nonlinear.

strength: the load or strain that, when applied once, usually fractures a bone (its “ultimate strength”). Normal lamellar bone's fracture strength expressed as a strain of approximately 25,000 μstrain (c.v. ≈ 0.3), which correspond to a change in length of 2.5%, i.e., from 100% of its original length to 97.5% of that length under compression, or 102.5% of it under tension. That fracture strain corresponds to an ultimate or fracture stress of approximately 17,000 pounds per square inch, or approximate 120 MPa. The large difference revealed by in vivo strain studies between bone's fracture strain and the largest normally allowed bone strain in life (the MESm, ∼1000 μstrain in most adults) surprised many skeletal physiologists.

turnover: here the amount of bone or the fraction of it that is replaced by new bone.

vital biomechanics: a subfield of general biomechanics that concerns how and why biologic mechanisms respond to mechanical usage and loads, and other physical stimuli, to adapt structural tissues and organs to their mechanical usage to make them mechanically competent, and then keep them so for life.


  1. Top of page
  2. Abstract
  8. References
  • 1
    BilezikianJP, RaiszLG, RodanGA, (eds.) 1996 Principles of Bone Biology. Academic Press, New York, NY, U.S.A.
  • 2
    Buckwalter JA, Woo SL-Y, Goldberg VM, Hadley EC, Booth F, Oregema TR, Eyre DR 1993 Soft tissue aging and musculoskeletal function J Bone Joint Surg 75A:15331548.
  • 3
    FavusMJ, (ed.) 1996 Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 3rd Ed. Lippincott-Raven Press, New York, NY, U.S.A.
  • 4
    Frost HM, 1995 Introduction to a New Skeletal Physiology, Vols I, II. The Pajaro Group, Inc., Pueblo, CO, U.S.A.
  • 5
    Jee WSS, Frost HM 1992 Skeletal adaptations during growth Triangle (Ciba Geigy) 31:7788.
  • 6
    Martin RB, Burr DB, 1989 Structure, Function and Adaptation of Compact Bone. Raven Press, New York, NY, U.S.A.
  • 7
    Ferretti JL 1995 Perspectives of pQCT technology associated to biomech anical studies in skeletal research employing rat models Bone 17(Suppl):353364.
  • 8
    Ferretti JL, Frost HM, Gasser JA, High WB, Jee WSS, Jerome C, Mosekilde L, Thompson DD 1995 Perspectives: On osteoporosis research: Its focus and some insights of a new paradigm Calcif Tissue Int 57:399404.
  • 9
    Abbasi-Jahromi SH, Matayoshi A, Kimble R, Dimarogonas A, Pacifici R 1996 Bone quality factor analysis: a new noninvasive technique for the measurement of bone density and bone strength J Bone Miner Res 11:594599.
  • 10
    Compston JE 1995 Bone density: BMC, BMD, or corrected BMD? Bone 16:57.
  • 11
    Gasser JA 1995 Assessing bone quantity by pQCT Bone 17(Suppl):145154.
  • 12
    Schiessl H, Ferretti JL, Tysarczyk-Niemeyer G, Willnecker J, 1996 Noninvasive bone strength index as analyzed by peripheral quantitative computed tomography (pQCT). In: SchönauE, (ed.) Paediatric Osteology: New Developments in Diagnostics and Therapy. Elsevier, Amsterdam, The Netherlands, pp. 141146.
  • 13
    Schönau E, Werhahn E, Schiedermaier U, Mokov E, Schiedhauer K, Rietschel E, Haverkamp F, Schiessl H, Michalk D, 1996 Bone and muscle development during childhood in health and disease. In: SchönauE, (ed.) Paediatric Osteology. New Developments in Diagnostics and Therapy. Elsevier, Amsterdam, The Netherlands, pp. 147160.
  • 14
    RiggsBL, MeltonLJIII, (eds.) 1995 Osteoporosis. Etiology, Diagnosis and Treatment, 2nd Ed. Lippincott-Raven Publishers, Hagerstown, MD, U.S.A.
  • 15
    Smith EL, Gilligan C 1989 Mechanical forces and bone J Bone Miner Res 6:139173.
  • 16
    Jee WSS, 1989 The skeletal tissues. In: WeissL, (ed.) Cell and Tissue Biology: A Text book of Histology. Urban and Schwartzenberg, Baltimore, MD, U.S.A., pp. 211259.
  • 17
    Frost HM, 1997 “Osteoporoses”: A Rationale for Further Definitions? Calcif Tissue Int (in press).
  • 18
    Turner CH, Takano Y, Owan I 1995 Aging changes mechanical loading thresholds for bone formation in rats J Bone Miner Res 10:15441549.
  • 19
    Forwood MR, Burr DB 1993 Physical activity and bone mass: Exercises in futility? Bone Miner 21:89112.
  • 20
    Jee WSS, Li XJ 1990 Adaptation of cancellous bone to overloading in the adult rat: A single photon absorptiometry and histomorphometry study Anat Rec 227:418426.
  • 21
    Jee WSS, Li XJ, Schaffler MB 1991 Adaptation of diaphyseal structure with aging and increased mechanical usage in the adult rat: A histomorphometrical and biomechanical study Anat Rec 230:332338.
  • 22
    Jee WSS, (ed.) 1995 Proceedings of the International Conference on Animal Models in the Prevention and Treatment of Osteopenia Bone 17(Suppl):1466.
  • 23
    Li XJ, Jee WSS 1991 Adaptation of diaphyseal structure to aging and decreased mechanical loading in the adult rat: A densitometric and histomorphometric study Anat Rec 229:291297.
  • 24
    Li XJ, Jee WSS, Chow S-Y, Woodbury DM 1990 Adaptation of cancellous bone to aging and immobilization in the rat: A single photon absorptiometry and histomorphometry study Anat Rec 227:1224.
  • 25
    Simon SR, Paul IL, Mansour J, Munro M, Abernathy PJ, Radin EL 1981 Peak dynamic force in human gait J Biomech 14:812822.
  • 26
    Crowninshield RD, Johnston RC, Andrews JG, Brand RA 1978 A biomechanical investigation of the human hip J Biomech 11:7585.
  • 27
    Simon SR, Radin EL 1993 Biomechanics of joints. In: KellyWNK, HarisED, RuddyS, SledgeCB, (eds.) Textbook of Rheumatology. WB Saunders Co., Philadelphia, PA, U.S.A., pp., 16751678.
  • 28
    Thompson D'Arcy W, 1917 On Growth and Form. University of Cambridge Press, Cambridge, U.K.
  • 29
    SchönauE, (ed.) 1996 Paediatric Osteology: New Developments in Diagnostics and Therapy. Elsevier Science, Amsterdam, The Netherlands.
  • 30
    TakahashiHE, (ed.) 1995 Spinal Disorders and Growth and Aging. Springer-Verlag, Tokyo, Japan.
  • 31
    Kannus P, Sievanen H, Vuori L 1996 Physical loading, exercise and bone Bone 18(Suppl 1):13.
  • 32
    Aniansson A, Zetterberg C, Hedberg M, Henriksson K 1984 Impaired muscle function with aging: A background factor in the incidence of fractures of the proximal end of the femur Clin Orthop 191:193201.
  • 33
    Faulkner JA, Brooks SV, Zerva E, 1990 Skeletal muscle weakness and fatigue in old age: Underlying mechanisms. In: CristofaloVJ, LawtonMP, (eds.) Annual Review of Gerontology and Geriatrics. Springer-Verlag, New York, NY, U.S.A., pp. 147166.
  • 34
    Kritz-Silverstein M, Barrett-Connor E 1994 Grip strength and bone mineral density in older women J Bone Miner Res 9:4551.
  • 35
    McCarter R 1990 Age-related changes in skeletal muscle function Aging 2:2738.
  • 36
    Murray M, Gardner G, Mollinger L, Sepic S 1980 Strength of isometric and isokinetic contractions: Knee muscles of men aged 20 to 86 Phys Ther 60:412419.
  • 37
    Raloff J 1996 Vanishing flesh: Muscle loss in the elderly finally gets some respect Sci News 150:9091.
  • 38
    Kalu DN 1995 Evolution of the pathogenesis of postmenopausal bone loss Bone 17(Suppl):135145.
  • 39
    Karlsson M, Johnell O, Obrant K 1993 Bone mineral density in weight lifters Calcif Tissue Int 52:212215.
  • 40
    Nilsson BE, Andersson SM, Hardtrup T, Westlin NE 1987 Ballet dancing and weight lifting—effects on BMC Am J Roentgen 131:541542.
  • 41
    Han Z-H, Palnitkar S, Rao DS, Nelson D, Parfitt AM 1997 Effects of ethnicity and age or menopause on the remodeling and turnover of iliac bone: Implications for mechanisms of bone loss J Bone Miner Res 12:498508.
  • 42
    Burr DB, Martin RB 1989 Errors in bone remodeling: Toward a unified theory of metabolic bone disease Am J Anat 186:131.
  • 43
    Frost HM 1987 The mechanostat: A proposed pathogenetic mechanism of osteoporoses and the bone mass effects of mechanical and nonmechanical agents Bone Miner 2:7385.
  • 44
    Frost HM, 1973 Bone Modeling and Skeletal Modeling Errors. Charles C. Thomas, Springfield, MA, U.S.A.
  • 45
    Burr DB, Milgrom C, Fyrhie D, Forwood M, Nyska M, Finestone A, Saiag E, Simkin A 1995 In vivo measurement of human tibial strains during vigorous activity Bone 18:405410.
  • 46
    Forwood MR, Turner CH 1995 Skeletal adaptations to mechanical usage: Results from tibial loading studies in rats Bone 17(Suppl):197205.
  • 47
    Lanyon LE 1996 Using functional loading to influence bone mass and architecture: Objectives, mechanisms, and relationship with estrogen of the mechanically adaptive process in bone Bone 18(Suppl 1):3743.
  • 48
    Nunamaker DM, Butterweck DM, Provost MP 1990 Fatigue fractures in thoroughbred race horses: Relationship with age, peak bone strain and running J Orthop Res 8:604611.
  • 49
    Turner CH, Forwood MR, 1995 Bone adaptation to mechanical forces in the rat tibia. In: OdgaardA, WeinansH, (eds.) Bone Structure And Remodeling. World Scientific, London, U.K., pp. 6578.