Bone biology is a complex and vastly growing area of study. It brings together the traditional fields of anatomy, physiology, and biomechanics with the increasingly complex fields of developmental biology and molecular genetics. For clinicians who treat bone disorders such as osteoporosis, developing a working knowledge of this topic is essential. This article discusses bone from a structural, anatomical, and functional perspective. It reviews skeletogenesis as a developmental process and from a regulatory perspective and presents biomechanical principles and theories. Osteoporosis is reviewed, including recent literature related to the role of exercise in prevention and treatment of this disease. [Downey PA, Siegel MI. Bone biology and the clinical implications for osteoporosis.]
The purpose of this perspective is to present physical therapists with a background on bone biology that can help them understand bone pathologies such as osteoporosis. It discusses the cellular and extracellular composition of bone as well as the process of bone formation, mineralization, and resorption. Furthermore, this article introduces readers to basic principles of genetic and molecular control of bone formation and reviews the biomechanical properties of bone. Lastly, this article links bone physiology with the underlying pathophysiology of osteoporosis and updates the literature regarding the role of exercise in prevention and treatment.
Bone is a specialized form of connective tissue that serves as both a tissue and an organ system within higher vertebrates. As such, its basic functions include locomotion, protection, and mineral homeostasis. Its cellular makeup includes osteoblasts, osteocytes, bone lining cells, and osteoclasts, and its matrix contains an organic and an inorganic component. Morphologically, bone is characterized either as cancellous (spongy, trabecular) or as cortical (compact). Functionally, cancellous bone is more closely associated with metabolic capabilities than cortical bone, whereas cortical bone generally provides greater mechanical strength. Although bone exhibits significant mechanical strength at a minimum weight, its biomechanical properties allow for significant flexibility without compromising this mechanical strength.
The 4 cellular elements of bone are: osteoblasts, osteocytes, bone lining cells, and osteoclasts.1,2 A simpler cellular classification consisting of bone forming and bone resorbing cells also has been developed.3 Further differentiation of bone cells is based on their origin. Osteoblasts, osteocytes, and bone lining cells originate from mesenchymal stem cells known as osteoprogenitor cells, whereas osteoclasts originate from hemopoietic stem cells. The location of these cells also varies. Bone cells found along the surface of bone include osteoblasts, osteoclasts, and bone lining cells, whereas osteocytes are located in the interior of bone (Fig. 1).1,4
As previously indicated, osteoblasts are derived from undifferentiated mesenchymal cells that are located in the marrow, endosteum, periosteum, and bone canals. These cells, also referred to as “preosteoblasts,” can migrate from surrounding tissue or through the vascular system. Mesenchymal cells are stellate in shape, contain relatively small amounts of cytoplasm and organelles, and possess a single nucleus. Differentiation and proliferation of mesenchymal cells into osteoblasts occurs during both intramembranous and endochondral bone formation.1,3
With the advent of electron microscopy, the structure of the osteoblast has become more defined. These robust cells are tightly packed along the surface linings of bone. When active, osteoblasts are oval and contain large quantities of rough endoplasmic reticula (RER), mitochondria, and Golgi apparatus. Their single nucleus is found within the center of the cell. Other microscopic components found within these cells include mitochondria, microtubules, microfilaments, lysosomes, glycogen, and lipids. Functionally, the osteoblast is responsible for production of the organic matrix, which is composed of proteins and polysaccharides.5 Evidence exists that osteoblasts, under the influence of parathyroid hormone and local cytokines, release mediators that activate osteoclasts.1
Eventually, osteoblasts follow 1 of 3 pathways. These cells may (1) remain active osteoblasts, (2) become surrounded by matrix and become osteocytes, or (3) become relatively inactive and form bone lining cells. Bone lining cells are thin, elongated cells that cover most bone surfaces in the mature skeleton. Cytoplasmic extensions or gap junctions often link them to each other or to osteocytes. Because they are metabolically inactive, bone lining cells contain fewer organelles and less cytoplasm than osteoblasts. At times, they are referred to as “resting osteoblasts” or “surface osteocytes.”1,2,4,5 Researchers still speculate about the function of these cells. Buckwalter et al1 indicated that, in the presence of parathyroid hormone, these cells secrete enzymes that remove the osteoid covering of the bone matrix in preparation for osteoclastic removal of bone. Other authors2,4 reported that bone lining cells may be precursors for osteoblasts, regulate the crystal growth in bone, or function as a barrier between extracellular fluid and bone.
The third cell type, the osteocyte, is estimated to make up more than 90% of the bone cells in an adult skeleton. As immature osteocytes, recently surrounded in bone matrix, they closely resemble osteoblasts. Thus, the cytoplasm contains large amounts of RER and large Golgi apparatus and mitochondria, with lesser amounts of microtubules, microfilaments, and lysosomes. As these cells mature and more matrix is laid down, osteocytes become located deeper within the bone tissue and eventually become smaller as they lose cytoplasm. This accounts for the enlarged appearance of their nucleus. Furthermore, they are located within a space or lacuna and have long cytoplasmic processes that project through canaliculi within the matrix and that contact processes of adjacent cells. These connecting processes are thought to be extremely important in cellular communication and nutrition within a mineralized matrix.2,4,5 Moreover, this important cellular network is thought to allow cell-mediated exchanges of minerals between the fluids in the bone and the vascular supply. It also is believed that the cellular network senses the mechanical deformation within bone that leads to the coordinated formation and resorption of bone.1
The fourth cellular component of bone is the osteoclast. This giant, multinucleated cell is responsible for bone resorption under both normal and pathological conditions, such as osteoporosis. Morphologically, osteoclasts tend to be much larger than other bone cells and are generally located on the surface of bone. They are known to be very mobile, moving from various sites and along the bone surface, and this motility is thought to account for the varied appearance of these cells.5
Osteoclast nuclei, which average between 3 and 20, tend to be oval and concentrated mid-cell. There is less RER present than in osteoblasts, which is consistent with decreased production and secretion of proteins. Mitochondria are more numerous within osteoclasts than any other cell type within the body. Between the nuclei are vesicles of Golgi material, which are relatively small in number. Many lysosomal types of vacuoles are present, leading to the common description of the cytoplasm as being “foamy.”1,6,7 The plasma membrane of the active osteoclast has an infolded appearance known as a ruffled border. The deep infolds of this border result in appendage-like projections of the cell that can wrap around bony prominences or lie along the surface. The large membrane surface area potentially permits extensive exchange between the intracellular and extracellular environments.1,6
The extracellular makeup of bone comprises approximately 90% of its volume compared with the remaining 10% comprising cells and blood vessels. This extracellular matrix is composed of both an organic component and an inorganic component. The organic matrix accounts for approximately 35% of the total weight of bone tissue compared with 65% for the inorganic part.1
The organic makeup of bone consists primarily of collagen that is synthesized by osteoblasts, secreted, and then assembled extracellularly. Type I collagen predominates, but types V, VI, VIII, and XII are present in small amounts. Intracellularly, type I collagen molecules, which are made up of a triple helix of polypeptide chains, form collagen fibrils extracellularly. These fibrils are precisely arranged in an overlapping manner, maintaining spaces between adjacent fibrils. Numerous intermolecular crosslinks are formed, producing a stable, porous structure from which the bone, in part, derives its ultimate yield strength.7,8 This is analogous to the plastic region of a stress-strain curve for bone as illustrated in Figure 2.8
The inorganic matrix of bone is essential in providing the major portion of the tensile yield strength and the important physiological functions related to the storage of ions. It is estimated that the mineral salts of bone contain 99% of the calcium, 85% of the phosphorus, and between 40% and 60% sodium and magnesium found in the body. The physiological functions related to nerve conduction and muscle contraction depend on this inorganic matrix in order to maintain appropriate extracellular fluid ion concentrations.
Bone mineral crystals, previously thought to be classified as pure hydroxyapatite,9 are now regarded as apatite due to unique acid-phosphate groups. The mechanical and physiological roles of bone mineral crystals appear to depend on the amount of crystal present as well as on the age of the crystal. Both of these factors are important when dealing with bone diseases and fracture healing1 and will be discussed later in this perspective.
During initial bone formation, the mineralization process is complex and well regulated, and it occurs quickly once initiated. At least 60% of the process occurs within hours, while the remaining mineralization gradually increases the density and strength of bone. Exactly how the process is controlled is still under investigation, but it is believed that a variety of mechanisms related to the minor organic components of the matrix and their precursors, rather than the collagen itself, are responsible.10
Calcium phosphate granules are found within an osteoblast’s mitochondria and matrix vesicles, which are membrane-bound extracellular structures formed from the plasma membrane of osteoblasts. In the case of mineralization of the epiphyseal plate, they are formed from the plasma membrane of the chondrocyte. The minerals are deposited within and between adjacent collagen fibrils, and the crystals are aligned in a parallel fashion. As this process proceeds, the amount of water and noncollagenous proteins decrease. Although the concentration of minerals increases, the organization and amount of collagen remains the same. This correlates with the increasing strength and stiffness of bone.1,2
As relevant research is being published, especially in the area of osteoporosis, the process of bone resorption is becoming better understood. The process is initiated by the proliferation of osteoclast precursors, their differentiation into osteoclasts, followed by the degradation process of the bone matrix. Initially, the inorganic matrix dissolves through acidication (HCl) of the extracellular environment, which causes a decrease in pH. Following this, cathepsin K, a lysosomal protease, degrades the organic component. By-products of this process are then transported to the opposite side of the osteoclast where they are released. Following the degradation of the bony matrix, a “resorptive lacuna” is left behind. The osteoclast will detach from the site and move on potentially to a new site of resorption. Osteoblasts will then come into the area and replace the resorbed bone.11
During the adolescent years, bone density increases rapidly, reaching a maximum approximately 10 years after the completion of skeletal growth. In the aging person, the balance between resorption and deposition becomes negative, because the amount of resorption exceeds the deposition. In the 10 years following menopause, women lose close to half of their cancellous bone and one third of their cortical bone mass because of accelerated bone loss from estrogen depletion. In contrast, men lose approximately 30% less bone mass during their lifetime. Once this imbalance has become clinically significant, a person is diagnosed with osteoporosis.12
The overall structure of bone can be divided into cortical (compact) versus cancellous (spongy) (Fig. 3). Within these classifications, cortical and cancellous bone can consist of either woven (primary) or lamellar (secondary) bone. Comparison of cortical and cancellous bone demonstrates a similar matrix structure and composition, but vastly different masses, with cortical bone having a greater mass-to-volume ratio.1
Cortical bone surrounds the marrow cavity and the trabecular plates of the cancellous bone. It accounts for 80% of the mature skeleton and forms the diaphysis, or shaft, of long bones. The metaphysis and epiphysis of long bones have thinner cortical walls, with the epiphysis forming a bulbous end surrounding the inner cancellous bone. Short bones (eg, the tarsals and carpals), the vertebrae, skull, and pelvic bones also tend to have thinner cortical walls but contain a greater percentage of cancellous bone compared with long bones.1,13
The differences in mechanical properties between cortical and cancellous bone are due to the differences in architecture, even though the composition and materials are the same. The thick, dense arrangement of the diaphysis of long bones allows cortical bone to have a much higher resistance to torsional and bending forces, whereas cancellous bone provides greater resilience and shock absorption, such as in the epiphyseal region of long bones. Cancellous bone generally has a higher metabolic rate and appears to respond quicker to changes in mechanical loading and unloading, such as seen with prolonged immobilization. This may be due, in part, to the greater exposure of bone cells within cancellous bone to the adjacent bone marrow cells and vascular supply, whereas cells within cortical bone tend to be embedded deeper within the bone matrix.1
Cortical and cancellous bone can be made up of either woven or lamellar bone. Woven bone, sometimes referred to as primary bone, is seen in embryonic bone that is later resorbed and replaced by lamellar, or secondary, bone by 4 to 5 years of age. Woven bone, however, also is seen during the initial stages of fracture healing, within cranial sutures, ear ossicles, and epiphyseal plates. Exemplified by the relatively quick turnover rate during deposition and resorption, woven bone has a greater rate of metabolic activity compared with lamellar bone. Woven bone has a scattered, irregular appearance, whereas lamellar bone has a very orderly arrangement. Histologically, the osteocytes seen in woven bone also are more randomly scattered than those in lamellar bone, whereas the osteocytes are uniform in size and shape and are oriented in line with the other cells and structures within the bone.9
When lamellar bone is viewed microscopically in cross-section, the organization of the layers appears in parallel units or sheets with densely packed collagen fibrils. Concentric rings of lamellae form osteons, which are also known as haversian systems. Osteons surround central canals (haversian canals), which contain blood, lymph vessels, and, occasionally, nerves. Between the central canals and the surrounding cells are the cell processes of osteocytes, which travel within tunnel-like structures known as canaliculi. They extend out in a radial manner between the central canals and surrounding osteocytes (Fig. 4). This allows for diffusion of nutrients in a system that is surrounded by a hard, mineralized matrix. The central canals also branch and anastomose with obliquely oriented vascular branches known as Volkmann canals. These structures allow for extended communication from the periosteum to the endosteum.12
Primary osteons undergo resorption and new osteons form, leaving behind boundaries known as cement lines. The constant resorption and deposition of new bone is the basis for the dynamic process of bone turnover. Histologically, it is possible to see areas within a cross-section of bone where remnants of primary osteons exist along with secondary osteons.12
The complex and dynamic network of lacunae and canals within bony tissue form an extravascular space where, adjacent to a mineralized matrix, fluids and ions can flow relatively unrestricted, and mechanical bone deformations can be converted to electrical signals and transmitted to other areas of the tissue. Some authors8,14 have hypothesized about the role of electrical signals in the regulation of bone function based on this interdependent network. This idea will be further discussed in the biomechanical section of this perspective.
Skeletogenesis is the process through which bone is laid down to form eventually a mature skeleton. Technically, there is only one mechanism of bone formation: the laying down of the osteoid matrix by osteoblasts, followed by the deposition of crystalline apatite. However, there are 2 different methods of the ossification process: intramembranous ossification is bone formation from an organic matrix membrane, whereas endochondral ossification occurs within a cartilaginous model. Intramembranous ossification accounts for the formation of the vault of the skull and most of the mandible and clavicle; endochondral ossification occurs with the axial and appendicular skeleton along with the base of the skull12,14 and parts of the mandible15 and will be discussed below in more detail.
Endochondral Bone Formation
Endochondral bone formation has simplistically been referred to as replacement of cartilage by bone, but this process is very complex in both its molecular and cellular transitions.16 The cartilaginous model for this type of bone formation, as indicated previously, is derived from mesenchymal cells.
Mesenchymal, or prechondrogenic, cells are similar to fibroblasts in appearance and in their ability to synthesize collagen, fibronectin, and noncartilage-type proteoglycans. The beginning of the cartilage differentiation process is signaled by cellular condensation of the mesenchyme prior to cartilage matrix secretion. The mesenchymal cells differentiate into chondroblasts, which proliferate and produce a matrix that forms both the shape and position of the eventual bone. The embryonic model for long bones consists of hyaline cartilage, which undergoes appositional growth resulting in a dumbbell-like shape. The shaft of the cartilaginous mass becomes the diaphysis, with the epiphyses located at both ends, and completely surrounded by the perichondrium.7 In the central aspect of the forming bone, long linear columns of chondrocytes progressively hypertrophy, resorb the surrounding cartilage, and leave behind trabeculae of cartilage matrix, which then becomes mineralized. The chondrocytes degenerate, leaving behind interconnected spaces.4,7
Along the circumference of the developing endochondral bone, the perichondrium develops osteogenic potential and lays down a thin layer of bone around the shaft, known as the periosteum. Primitive mesenchymal cells and blood vessels then invade the spaces within the shaft of the bone that are left after the chondrocytes degenerate. This mesenchyme differentiates into osteoblasts and bone marrow cells.4,7 Irregular woven bone is then formed as the osteoblasts layer along the surface of the calcified cartilage remnants. The ends of the original cartilaginous model are now separated through this process known as primary ossification.7,17
Secondary ossification also is occurring within the epiphyses, while a thin layer of hyaline cartilage is retained along the articular surface. The border between the diaphysis and epiphysis of developing long bones is the epiphyseal or growth plate. This is the area of continued longitudinal bone growth until physical maturity, when cartilage is replaced by bone, bringing together the diaphysis and the epiphysis.7
Histologically, the cartilaginous zone of the growth plate is divided based on the function or morphology of the chondrocytes. The reserve zone is located furthest away from the diaphysis, followed by the proliferative and hypertrophic zones. The reserve zone contains cells that undergo little to no cell division and function as stem cells for the adjacent proliferative zone. The proliferative zone, recognized by long columns of flattened cells, secretes extracellular matrix and undergoes rapid cell division. The hypertrophic zone is responsible for longitudinal bone growth (Fig. 5). A fourth region, previously identified as the zone of degeneration, is now recognized as an artifact.18,19
Growth plate activity leads to the continual production of cellular and cartilaginous matrix and results in longitudinal bone growth. The growth plate itself, however, does not increase in size, because there is a continuous balance of tissue resorption and deposition at the epiphyseal/metaphyseal junction (Fig. 6).9,18 Injuries to the growth plate can result in disruption of blood supply, which may result in necrosis of the plate and cessation of growth. Growth disturbances also can result from fractures that extend through the epiphyseal plate and are classified commonly as Salter-Harris fractures. A Salter-Harris type IV fracture, an intra-articular fracture that extends from the epiphysis through the plate and into the metaphysis, along with a type V fracture, which crushes one side of the plate, can result in arrested growth.20
Genetic and Molecular Regulation of Bone Development and Remodeling
Bone development, including cellular differentiation, growth, and repair, is controlled by genetic and epigenetic factors.21 Differential gene activity, the turning on and off of genes at various times during bone development, regulates cell differentiation and ultimately the morphogenesis of bone. Genes produce transcription products that are translated into regulatory, enzymatic, or structural proteins. The transcription products provide the blueprint for thousands of proteins that eventually interact through signal transductions to influence cellular differentiation, such as mesenchymal cells differentiating into chondroblasts or osteoblasts.22
Epigenesis is the developmental process whereby an organism grows from a simple to more complex form through progressive differentiation of undifferentiated cells. This process occurs following genetic determination and includes regulation by systemic and local factors.22 Common local factors include cytokines, growth factors, and prostaglandins. Hormones, including parathyroid, vitamin D3, and calcitonin are important systemic factors involved in bone development.12
Parathyroid hormone (PTH), secreted by the parathyroid gland, is known to regulate calcium homeostasis by increasing the release of calcium from bone and the resorption of calcium by the kidneys.12,23 Parathyroid hormone is known to be a strong stimulator of osteoclastic bone resorption. It has been shown to stimulate pre-existing osteoclasts, increase the number of osteoclasts with active ruffled borders, and expand the ruffled borders within individual osteoclasts. The changes seen in the numbers of osteoclasts and their level of activity parallel the increase seen in extracellular calcium.2 Evidence also supports the role PTH plays in causing bone lining cells to retract from the mineralized osteoid, thereby providing the osteoclast with a physical space to attach to the matrix.12
The active form of vitamin D, 1,25-dihydroxyvitamin D3, although it has several actions, is primarily related to bone metabolism and mineral homeostasis. Both inhibition and induction of osteoblastic activity have been demonstrated at the cellular level, depending on whether the vitamin D is applied during the proliferative or differentiation stages of development.4,24 Vitamin D also plays an important role in enhancing calcium absorption in the intestine and inhibiting PTH synthesis and secretion.25 Although no direct role between vitamin D and bone mineralization has been established, insufficient levels of this vitamin are associated with the childhood disease of rickets, which results in decreased bone mineralization. Serum calcium and phosphorus levels are insufficient to support mineralization, but dietary supplementation of vitamin D will generally correct the imbalance.2
Calcitonin, a polypeptide hormone synthesized by the thyroid gland, has a significant inhibitory effect on osteoclasts, thereby lowering the levels of serum calcium. The osteoclast appears to be a main target of calcitonin even at low levels of concentration. Within 30 minutes of administration of therapeutic, pharmacological doses of calcitonin, a complete inhibition of osteoclastic bone resorption occurs, accompanied by the loss of ruffled borders, loss of cytoplasm along the ruffled border, and a physical dislocation from the underlying bone.12,26 Calcitonin has been used therapeutically in the treatment of Paget disease and osteoporosis. However, concerns over the calcitonin-induced loss of calcitonin receptors, which results in a hormone induced resistance, has led to concerns about its long-term use in disease treatment or prevention.27
Other hormones that influence bone cell function include glucocorticoids, thyroid hormone, and estrogens. Glucocorticoids have several complex effects on bone metabolism, the best known of which is an inhibitory effect on the osteoblast’s ability to synthesize bone matrix. Prolonged use of glucocorticoids can result in osteopenia. Osteopenia, moreover, also can result from the thyroid hormones thyroxine and tri-iodothyronine, both of which act to stimulate osteoclastic resorption of bone. Estrogens have several, complex effects on bone cell function including effects on calcitonin, PTH, and vitamin D. Estrogens overall appear to decrease the rate of bone turnover, specifically by influencing osteoclastic activity. Estrogen deprivation results in an increase in bone remodeling sites with a possible long-term result of osteoporosis.12 Clinically, accelerated bone loss can be seen in the postmenopausal female population and will be discussed later in more detail.
Unlike hormones that regulate bone development through systemic mechanisms, local factors (eg, cytokines, growth factors, prostaglandins) influence development by cell-to-cell and cell-to-extracellular matrix interactions. Cytokines and growth factors are soluble molecules that act at a local level, mediating cell-to-cell interactions within bone. Their regulatory function begins with growth and development and continues in the mature skeleton through the remodeling process. Prostaglandins are a diverse group of unsaturated fatty acids that are thought to be able to regulate a variety of processes, including inflammation, blood flow, and ion transport across membranes. Initially, they appear to have an inhibitory effect on osteoclasts, but subsequently have a stimulatory effect on bone resorption by increasing formation and proliferation of osteoclasts. Prostaglandin E has been a factor associated with the bone loss seen in disease processes such as rheumatoid arthritis, periodontal disease, and possibly neoplasms.12,28
Because bone is a living, dynamic connective tissue, it provides mechanical support related to protection and locomotion, and it functions as a system of complex metabolic mineral homeostasis. In contrast to other nonbiologic materials, bone demonstrates the mechanical properties of anistrophy, nonlinearity, and viscoelasticity. These properties, along with its ability to respond to changes in its physiological and mechanical environment, make it more difficult to establish universal constants related to the physical properties of bone.8 This section will discuss some of the basic physical properties of bone and then review selected theories related to growth and remodeling.
Based on biomechanical principles, bone responds to forces in nature, including gravity, ground reaction, and muscle contraction. When a force or a load is applied to bone, an internal resistance develops (ie, stress). Stress is the force per unit area and is equal in magnitude but opposite in direction to the applied load. Stress can be categorized as (1) tensile, occurring when 2 forces act along a straight line in opposite directions; (2) compressive, occurring when 2 forces act along a straight line in the same direction; or (3) shear, occurring when 2 forces are acting parallel to each other but not in the same line. Most forces applied to bone are a combination of the 3 stresses, resulting in a bending or torsion. The resulting deformation of the applied force is known as strain, which is equal to the change in length divided by the original length.18,29
At low levels of stress, a linear relationship exists between stress and strain. The ratio of the stress divided by the strain is known as the modulus of elasticity or Young’s modulus. This relationship or modulus relates to the overall stiffness or rigidity of bone. The linear portion of the stress-strain curve is known as the elastic region, where removal of the load results in no permanent strain or deformation. The point at which the curve becomes nonlinear, the plastic region, a permanent deformation occurs even after the load is removed. This occurs at the elastic limit or yield point. Stressing a bone beyond the plastic region will result in failure, such as a fracture. The ultimate strength of a bone is determined by calculating the maximum stress at the point of failure (Fig. 2).8,29
Nonbiological materials often demonstrate the property of isotrophy, which means that, regardless of the direction of stress, the mechanical properties of the material will respond in the same manner. Bone, like many other biological tissues, demonstrates the property of anisotropy; it responds differently depending on the type of load applied. Cortical bone has been shown to resist compressive forces better than tensile forces. Compared with cortical bone, cancellous bone has a lower modulus of elasticity due to its greater porosity. Cancellous bone demonstrates the greatest strength when a compressive force is applied parallel to the trabecular system, such as a vertical force to a vertebral bone. Therefore, the strength and rigidity of bone are greatest in the direction of normal loading. Bone also demonstrates the property of viscoelasticity, which indicates that materials will demonstrate different properties according to the rate of force application. At low rates of loading, bone demonstrates a lower modulus of elasticity, and behaves like a viscous material. At higher rates of loading, bone behaves as a brittle material.8
Mechanical loads applied to bone are thought to be communicated through the bone by way of a mechanical signal detected by either bone lining cells or osteocytes, or both. It is believed that these mechanical signals lead to the generation of chemical signals involved in the regulation of bone formation and remodeling. The osteocytes, in particular, have received much attention in this regard. Osteocytes are connected to each other and to osteoblasts by way of cellular processes within canaliculi and are linked by gap junctions. This network allows for the possibility of electrical coupling as well as intracellular and extracellular molecular transport in cells deep within bone tissue.8
The functional relationship between the mechanical stresses and the structure of bone has been studied formally since 1892 in terms of Wolff’s law. Wolff’s law states that bone adapts during its growth to the functional forces acting upon it. The law indicates that there is a correlation between the direction of the principal stresses during normal function and the resulting pattern of trabecular alignment. The realization of this principle is seen in the femur, where the trabecular orientation corresponds to the directions of stress.8,30 Conversely, bone resorption has been shown to exceed deposition in the case of prolonged immobilization.31 Although the basic principles of Wolff’s law are widely accepted today, it is limited in its ability to explain all the biomechanical principles of bone development and remodeling.
Sperber14 stated that the basic shape and size of bone have a genetic determination. Once the morphology is established, relatively minor environmental features, such as bony tuberosities, develop. Nutritional, hormonal, and functional influences affect bone, and because osseous tissue is continually replaced throughout life, it will morphologically respond to mechanical stress. Sperber described 3 classifications of morphological features based on the influence of muscle. These features include those that develop only when muscle is present (temporal and nuchal lines), those that develop but require the presence of muscle to persist (angle of the mandible), and those that are associated with muscle but are mostly independent of its influence (body of the mandible and zygomatic bone).
The precise mechanisms by which mechanical forces influence bone structure and development are not known. The mediation of mechanical stress through piezoelectric currents has been postulated as having an influence in this area. Bone is made up of crystalline matrix, which enables it to generate small electrical currents in response to mechanical deformation. Therefore, it is hypothesized that the cellular makeup of bone may react to the electrical fields by laying down new bone.14 Clinically, we have seen electrical stimulation play a role in bone healing since as early as 1812 in the treatment of a tibial nonunion. Since that time, bone stimulators have demonstrated efficacy in augmenting open reduction surgeries either with internal or external fixation and bone grafts and in helping to treat infected nonunions and failed arthodeses.32 Methods for delivering electrical stimulation for bone healing include direct current applied either percutaneously or implanted at the fracture site and external application of either capacitive coupling or pulsed electromagnetic field. Although the exact mechanism of electrical stimulation that induces bone healing is not clear, electromagnetic fields have been found to stimulate the production of transforming growth factor and bone morphogentic proteins, both of which are involved in osteogenesis.33
A second theory related to the influence of biomechanical forces on bone tissue is based on the mechanochemical hypothesis, whereby the loads applied to bone are translated into cellular activity through straining of apatite crystals, altering the solubility of apatite, and changing local calcium concentrations. This process either stimulates or resorbs bone.14 Russell et al34 also discussed the influence of early biochemical responses to mechanical loading, resulting in the induction of prostacyclin synthesis, an increase of insulin-like growth factors, and changes in amino acid transporters, ultimately resulting in new bone formation.
One of the most interesting applied areas of bone biology for physical therapists is that of osteoporosis. Osteoporosis is a relatively common clinical disorder in which the process of bone resorption is increased. It disproportionately affects women more than men and is estimated to affect 1 in 3 women beyond the age of 50 years. It has been projected that approximately 9.4 million women in the United States have lost more than 25% of their peak bone mass,27,34 and 1.3 million fractures occur annually secondary to osteoporosis.35
Osteoporosis is a condition of microarchitectural loss of bone tissue leading to decreased density and bone fragility (Fig. 7). The primary reasons for developing this condition include poor bone acquisition during youth and accelerated bone loss during aging. Both of these processes are regulated by environmental and genetic controls. Loss of bone mass can be due to a combination of hormone deficiency, poor nutrition, decreased physical activity, and various pharmacological agents.35 One possible hypothesis for the pathogenesis of osteoporosis is that the complex system that controls local adaptation to mechanical stress is impaired during the normal aging process. Changes may occur in the production of local factors that mediate the response to mechanical stress. The loss of functional loading in the elderly and its ultimate role in the pathogenesis of osteoporosis also is not fully resolved at this time.36
Osteoporosis can be categorized as being either primary or secondary. Primary osteoporosis is the deterioration of bone mass associated with either a decrease in sex hormone, aging, or both. In women, early menopause or premenopausal estrogen deficiencies can accelerate the development of primary osteoporosis (Fig. 8). Secondary osteoporosis can occur due to chronic conditions that contribute to the acceleration of bone loss, including excess endogenous and exogenous thyroxin, malignancies, gastrointestinal diseases, hyperparathyroidism, connective tissue diseases, renal failure, and medications.35 The most common medication-related inducement of osteoporosis is long-term use of glucocorticoids, which is associated with suppression of osteoblastic activity. Other contributing factors include prolonged periods of inactivity or immobilization, inadequate calcium intake, and alcohol and tobacco abuse.27
Risk factors for developing osteoporosis include genetic, nutritional, and behavioral. Genetic factors include female sex, a petite skeletal frame, and Caucasian or Asian ancestry. Low calcium or vitamin D intake, alcohol abuse, and high caffeine intake are nutritional factors, and sedentary lifestyle, nulliparity, aging, smoking, and low body weight are some of the behavioral risk factors.35
The most common cause of osteoporosis is the decrease in the female sex hormone, estrogen, which occurs following menopause. An increase in bone resorption, which is associated with a rise in the number of osteoclasts, is correlated with the loss of estrogen. This increase in osteoclasts is caused by an increase in the cytokines that regulate the production of osteoclasts. It is believed that estrogen, either directly or indirectly, regulates the production of these cytokines.36
The goal of osteoporosis therapies is to inhibit bone resorption. This is achieved by reducing osteoclastic production or activity. Common pharmaceutical therapies that physical therapists may see their patients take include estrogens and selective estrogen receptor modulators (SERMs), bisphosphonates (BPs), and calcitonin.27
Hormone replacement therapy (HRT) has been shown to inhibit bone loss and bone turnover and actually increase bone mineral density (BMD). The molecular mechanism of action of estrogen on bone is poorly understood. Estrogen receptors have been identified, but their contribution to the total effect of this therapy is still being investigated. Due to the increased risk of endometrial cancer, which is associated with estrogen replacement therapy, progestin has been given in combination in the form of Premarin* for women with an intact uterus.27,37 However, recent results from the Women’s Health Initiative, the first randomized primary prevention trial for postmenopausal hormones, has demonstrated an increased risk of breast cancer, coronary heart disease, and pulmonary embolism in a small percentage of women who are taking estrogen/progestin HRT.37 The risks associated with HRT, therefore, must be weighed against the benefits in determining the best treatment for osteoporosis.
The SERMs exert estrogen-agonist effects on selective tissues. The most common of these are tamoxifen (Nolvadex†) and raloxifene (Evista‡). The mechanism exerted to inhibit bone resorption appears to be similar to that of estrogen—the blockage of cytokine production and, therefore, osteoclast differentiation.27
Bisphosphonates, the most common of which are etidronate (Didronel§) and alendronate (Fosamax‖), exert their effects on bone by inhibiting resorption. They have been shown to increase bone mass and reduce fracture rates of the spine and hip by 50% in women who are postmenopausal.34 Calcitonin, although used for many more years than BPs, appears to play less of a role in the treatment of osteoporosis. Calcitonin is a polypeptide hormone that also inhibits resorption by blocking osteoclastic activity. A negative side effect, the loss of calcitonin receptors, results in an overall hormone-induced resistance that has led many medical providers to choose other interventions.27
For physical therapists, understanding the role of exercise, whether it is weight-bearing exercise or exercise that focuses on improving the force-generating capacity of muscle, is vital in the prevention and treatment of osteoporosis. As mentioned previously, mechanical loads applied to bones create strain and the larger the load, the greater the strain. This strain is transmitted to the bone cells (osteoblasts, bone lining cells, and osteocytes), which are well suited to sense load changes due to their physical connections. Bone research has demonstrated that, in response to mechanical strain, there is an increase in cell metabolism and collagen synthesis.38 Animal studies have demonstrated dose-response relationships between loading and bone formation. Within 5 days of a single loading session, bone lining cells transformed into active osteoblasts. Four-point bending studies of bone also have demonstrated increased cell metabolism and proliferation in the tibial periosteum of rats.39 When the load exceeds the threshold for modeling or remodeling, as is possible with weight-bearing or weight-lifting exercise, bone strength gradually increases. Research related to the role of exercise in the treatment or prevention of osteoporosis has significantly improved our understanding of this phenomenon.
Chow et al40 compared an aerobic exercise group, an aerobic and strengthening exercise group, and a control group in a year-long, randomized controlled trial of 48 postmenopausal women between the ages of 50 and 62 years. The authors demonstrated a significant difference in total bone mass of the exercising groups compared with the control group, but no difference was found between the 2 different exercise groups. The aerobic exercise consisted of 30 minutes of walking, jogging, and dance, while the aerobic and strength training group had an additional 10 to 15 minutes of “low-intensity” isometric and isotonic exercise for the trunk and limbs, using free weights on the ankles and wrists for a 10-repetition maximum exercise protocol.
In a study of 124 postmenopausal women between 50 and 70 years of age, Bravo et al41 compared a group that received a combination of aerobic dancing, weight-bearing walking and stepping, and flexibility exercise with a control group. The authors demonstrated a significant decrease in spinal BMD in the control group and a stabilization of spinal BMD in the exercising group after 1 year. There was no change in femoral BMD, however, in either group.41
In contrast to the findings of Bravo et al,41 Prince et al42 looked at the effect of weight-bearing exercise on 6 different sites: lumbar spine, 3 hip sites, and 2 tibia sites. The experimental group, undergoing weight-bearing exercise with calcium supplementation, demonstrated a cessation of bone loss at the intertrochanteric hip site but at no other site including lumbar spine. The exercise program in this study included 2 hours of supervised exercise class and 2 hours of independent walking per week. Overall adherence with the exercise program, however, was assessed to be poor, with the exercise group performing an average of only 10% more activity compared with the nonexercising groups. Similarly, Lau et al43 demonstrated a significant effect on BMD at the femoral neck but not other areas of the femur or the lumbar spine when comparing a combination of load-bearing exercise and calcium supplementation.
Nelson et al44 noted differential effects on various skeletal sites when comparing a group receiving a 1-year walking program plus calcium supplementation with a sedentary group receiving calcium supplementation. The authors concluded that a walking program (4 times per week, 50-minute sessions) decreased bone loss of the spine compared with sedentary women who lost bone at this site regardless of dietary calcium intake. In contrast, high versus medium levels of calcium intake were shown to be a factor in reducing bone loss at the femoral neck.
Level of exercise intensity and impact also are factors that have been reported in the literature. A comparison of high- versus low-impact exercise on BMD of the lumbar spine of women who were postmenopausal demonstrated a decrease in bone density in the control group, whereas both exercise groups maintained BMD after a 1-year period. There was no significant difference between the group that performed high-impact exercise (exercise with peak forces at a minimum of 2 times body weight) and the group that performed low-impact exercise (exercise with peak forces at a maximum of 1.5 times body weight).45 In addition, a study was published comparing the effects on vertebral BMD of walking at a level below versus above an anaerobic threshold. The results demonstrated a decrease in lumbar vertebrae density in both the control group and the group exercising at a level below the anaerobic threshold, while the group exercising at a level above the anaerobic threshold had a significant increase in density.46
High-impact weight-bearing exercise also has been demonstrated to affect BMD in women who are premenopausal. Ninety-eight women who were healthy, but sedentary, and between the ages of 35 and 45 years were randomly assigned to a control group or to an experimental group. The experimental group participated in an 18-month progressive, high-impact exercise program. Significant differences in BMD at the femur were found between the groups, but no difference was demonstrated at non–weight-bearing sites, such as the distal radius.47
Weight training exercise, as a way to retard the effects of osteoporosis, also has been studied, with overall encouraging results. In a study by Nelson et al,48 40 women who were postmenopausal underwent a 2-day-per-week high-intensity strength training program (80% of a 1-repetition maximum or 16 on the Borg Rating of Perceived Exertion Scale) that included concentric/eccentric contractions of hip, knee, and back in extension along with lateral pull down and abdominal flexion using pneumatic resistance machines. Femoral neck and lumbar spine BMD increased in the exercise group and decreased in the control group. Dynamic balance also was shown to improve in the exercise group.48
Notelovitz et al49 studied women who were surgically menopausal (women who had undergone bilateral ovary removal, with or without a hysterectomy), all of whom received HRT, and divided them into a nonexercising control group and a group receiving variable-resistance circuit training. At the end of 1 year, the exercising group demonstrated significant increases in the spine, radius, and total body BMD compared with the control group. Smaller, but still significant, increases in regional (lumbar and trochanter) but not total BMD were reported by Lohman et al50 in premenopausal women following 18 months of resistance training. Differences in BMD were actually detected as early as 5 months in the lumbar spine and as early as 12 months in the trochanteric area.
Exercise protocols involving resistance training of specific muscles that attach to the lumbar spine also have been studied. Postmenopausal women participated in a 1-year psoas muscle-training program by performing 60 repetitions daily of sitting hip flexion with 5 kg of resistance.51 After statistically controlling for nonadherence, the authors reported a significant difference in BMD in the lumbar vertebrae between the control and exercising groups.51 Another study52 examined the effect of strengthening the back extensors on premenopausal women (28–39 years of age) by having the experimental group perform prone back extension with a backpack containing weights equivalent to 30% of their maximal isometric extensor strength. After 2 years of exercise, 5 times per week, the experimental group demonstrated significantly greater extensor muscle strength; however, there was no difference in BMD compared with the control group. The authors did note that the exercise might have been at a subthreshold level for influencing bone density or that this particular type of exercise may not be appropriate for treatment of osteoporosis.52
When clinicians review articles related to osteoporosis and exercise in order to prescribe exercise programs, it can be difficult to make clinical judgments based on what may appear as conflicting information. Systematic reviews of the literature can be valuable tools when developing exercise approaches to the prevention and treatment of osteoporosis. The Cochrane Database of Systematic Reviews published a review of exercise programs related to osteoporosis in postmenopausal women in 2003.53 The authors examined a total of 90 studies, 18 of which were randomized controlled trials that met the inclusion criteria for further review. They concluded, based on these 18 studies, that aerobic, weight-bearing, and resistance exercise were all effective in influencing BMD in the spines of postmenopausal women. Walking for exercise also was shown to be effective on both the BMD of the spine and hip, whereas aerobic exercise was effective in increasing BMD of the wrist. It should be noted, however, that aerobic exercise was not well defined in the Cochrane review.
The National Osteoporosis Society recommends exercise along a calcium-rich diet to prevent osteoporosis, and various drug therapies in the treatment of osteoporosis.54 The exercise component needs to be site specific, because the primary effect on bone is localized to the targeted site. The exercise stress needs to exceed the level to which the bone has adapted; and it is likely that bone will revert back to preactivity levels if the exercise ceases. The cycle of bone activation, resorption, and formation appears to take approximately 3 to 4 months; therefore, BMD changes should not be expected prior to that.54 Based on the current literature, both weight-bearing exercise in the form of walking, running, and aerobic dancing along with weight-training exercise have a role in the prevention and treatment of osteoporosis. Exercise adherence is an important factor to consider when interpreting the studies related to osteoporosis and exercise.
Bone biology is a scientific field that requires an understanding of complex, interrelated areas of study. Detailed knowledge of the anatomy, histology, and physiology of bone tissue has developed dramatically in the last few decades. Our understanding of the process of osteogenesis is constantly under revision, because our knowledge of the genetic and molecular mechanisms controlling bone cell differentiation and growth continues to expand. Much of the impetus behind current bone research is in order to understand the pathogenesis and treatment of diseases such as osteoporosis. Factors such as cytokines, prostaglandins, and mechanical loading, all of which influence and control the local formation and remodeling of bone, have potential to change the way osteoporosis is treated. Future studies are needed to help uncover the mechanobiological rules that help to govern bone response to mechanical loading.55 A better understanding of these rules will allow physical therapists to be more prescriptive with exercise and level of weight-bearing activity to capitalize on a bone’s inherent ability to remodel.
Both authors provided concept/idea/project design. Dr Downey provided writing. Dr Siegel provided project management and consultation (including review of manuscript before submission).
↵* Wyeth Pharmaceuticals, PO Box 9299, Philadelphia, PA 19101.
↵† AstraZenica Pharmaceuticals LP, 1800 Concord Pike, Wilmington, DE 19850-5437.
↵‡ Eli Lilly and Co, Lilly Corporate Center, Indianapolis, IN 46285.
↵§ Procter & Gamble Pharmaceuticals Inc, Health Care Research Center, 8700 Mason Montgomery Rd, Mason, OH 45040-9462.
↵‖ Merck & Co Inc, PO Box 4 WP39–206, West Point, PA 19486-0004.
- Physical Therapy