Physical Activity and Bone Health

SUMMARY Weight-bearing physical activity has beneficial effects on bone health across the age spectrum. Physical activities that generate relatively high-intensity loading forces, such as plyometrics, gymnastics, and high-intensity resistance training, augment bone mineral accrual in children and adolescents. Further, there is some evidence that exercise-induced gains in bone mass in children are maintained into adulthood, suggesting that physical activity habits during childhood may have long-lasting benefits on bone health. It is not yet possible to describe in detail an exercise program for children and adolescents that will optimize peak bone mass, because quantitative dose-response studies are lacking. However, evidence from multiple small randomized, controlled trials suggests that the following exercise prescription will augment bone mineral accrual in children and adolescents: Mode: impact activities, such as gymnastics, plyometrics, and jumping, and moderate intensity resistance training; participation in sports that involve running and jumping (soccer, basketball) is likely to be of benefit, but scientific evidence is lacking Intensity: high, in terms of bone-loading forces; for safety reasons, resistance training should be <60% of 1-repetition maximum (1RM) Frequency: at least 3 d·wk−1 Duration: 10–20 min (2 times per day or more may be more effective) During adulthood, the primary goal of physical activity should be to maintain bone mass. Whether adults can increase bone mineral density (BMD) through exercise training remains equivocal. When increases have been reported, it has been in response to relatively high intensity weight-bearing endurance or resistance exercise; gains in BMD do not appear to be preserved when the exercise is discontinued. Observational studies suggest that the age-related decline in BMD is attenuated, and the relative risk for fracture is reduced, in people who are physically active, even when the activity is not particularly vigorous. However, there have been no large randomized, controlled trials to confirm these observations, nor have there been adequate dose-response studies to determine the volume of physical activity required for such benefits. It is important to note that, although physical activity may counteract to some extent the aging-related decline in bone mass, there is currently no strong evidence that even vigorous physical activity attenuates the menopause-related loss of bone mineral in women. Thus, pharmacologic therapy for the prevention of osteoporosis may be indicated even for those postmenopausal women who are habitually physically active. Given the current state of knowledge from multiple small randomized, controlled trials and large observational studies, the following exercise prescription is recommended to help preserve bone health during adulthood: Mode: weight-bearing endurance activities (tennis; stair climbing; jogging, at least intermittently during walking), activities that involve jumping (volleyball, basketball), and resistance exercise (weight lifting) Intensity: moderate to high, in terms of bone-loading forces Frequency: weight-bearing endurance activities 3–5 times per week; resistance exercise 2–3 times per week Duration: 30–60 min·d−1 of a combination of weight-bearing endurance activities, activities that involve jumping, and resistance exercise that targets all major muscle groups It is not currently possible to easily quantify exercise intensity in terms of bone-loading forces, particularly for weight-bearing endurance activities. However, in general, the magnitude of bone-loading forces increases in parallel with increasing exercise intensity quantified by conventional methods (e.g., percent of maximal heart rate or percent of 1RM). The general recommendation that adults maintain a relatively high level of weight-bearing physical activity for bone health does not have an upper age limit, but as age increases so, too, does the need for ensuring that physical activities can be performed safely. In light of the rapid and profound effects of immobilization and bed rest on bone loss, and the poor prognosis for recovery of mineral after remobilization, even the frailest elderly should remain as physically active as their health permits to preserve skeletal integrity. Exercise programs for elderly women and men should include not only weight-bearing endurance and resistance activities aimed at preserving bone mass, but also activities designed to improve balance and prevent falls. Maintaining a vigorous level of physical activity across the lifespan should be viewed as an essential component of the prescription for achieving and maintaining good bone health. INTRODUCTION In Caucasian, postmenopausal women, osteoporosis is defined as a bone mineral density (BMD) value more than 2.5 standard deviations below the young adult mean value (52), with or without accompanying fractures. Whether the same criteria should apply to premenopausal women, women of other races, or men remains to be confirmed. In the U.S. and other developed countries the incidence of osteoporosis is increasing at rates faster than would be predicted by the increase in the proportion of aged individuals. Multiple vertebral fractures and, in particular, hip fractures have a devastating effect on functional abilities and quality of life. The mortality rate for elderly individuals in the first year following hip fracture is as high as 15–20% (105). Even with no change in current incidence rates, it has been estimated that the number of hip fractures will double to 2.6 million by the year 2025, with a greater percentage increase in men than in women (38). Because low BMD greatly elevates the risk of fractures with minimal trauma, as with a fall to the floor, strategies that maximize bone mass and/or reduce the risk of falling have the potential of reducing morbidity and mortality from osteoporotic fractures. Although bone mass can be increased through pharmacologic therapy, physical activity is the only intervention that can potentially both 1) increase bone mass and strength and 2) reduce the risk of falling in older populations. There exist other bone health issues associated with exercise, including the risk of stress fractures with high-volume training and the bone loss associated with amenorrhea. However, the focus of this position stand will be on the effectiveness of physical activity to reduce risk for osteoporotic fracture, without specific reference to nutritional or genetic influences. Well-known principles of exercise training apply to the effects of physical activity on bone. For example, overloading forces must be applied to bone to stimulate an adaptive response, and continued adaptation requires a progressively increasing overload. It is important to emphasize that the stimulus to bone is literally physical deformation of bone cells, rather than the metabolic or cardiovascular stresses typically associated with exercise (e.g., % VO2max). Physical deformation can be measured by strain gauges on the bone surface, but is more commonly estimated by such surrogate measures as ground-reaction forces engendered during weight-bearing activities. Muscle contraction forces in the absence of ground-reaction forces (e.g., swimming) may also stimulate bone formation, but this is more difficult to estimate. A factor that is unique to skeletal adaptations to training is the slow turnover of bone tissue. Because it takes 3–4 months for one remodeling cycle to complete the sequence of bone resorption, formation, and mineralization (85), a minimum of 6–8 months is required to achieve a new steady-state bone mass that is measurable. The most common outcome measure used to assess the effects of physical activity on bone mass in humans is BMD, which describes the amount of mineral measured per unit area or volume of bone tissue (51). Dual-energy x-ray absorptiometry (DXA) is the standard method of measuring areal BMD in clinical and research settings. The lumbar spine and proximal femur are the most common sites of measurement by DXA because they are prone to disabling osteoporotic fractures. Other methods of assessing risk for osteoporosis include computed tomography (CT) measurement of spine volumetric BMD, and ultrasonography of the calcaneus, which provides an index of bone stiffness. Ultrasonography is widely available, easy to perform, and does not involve exposure to ionizing radiation, but should be used only as a screening test. Currently, BMD is the best surrogate measure of bone strength in humans and BMD has been estimated to account for 60% or more of the variance in bone strength (20,125). However, studies of animals suggest that changes in BMD in response to mechanical stress underestimate the effects on bone strength. For example, 5–8% increases in BMD were associated with increases in bone strength of 64–87% (48,116). The size of bone has a significant contribution to bone strength because the resistance of bone to bending or torsional loading is exponentially related to its diameter; furthermore, bone size may continue to increase during adulthood (93). Because bone architecture (i.e., geometry) is an important determinant of strength (104), evaluation of the effects of mechanical stress on bone should consider not only changes in bone mass, but changes in structural strength and material and geometric properties when possible (120). The two generally accepted strategies to make the skeleton more resistant to fracture are to 1) maximize the gain in BMD in the first three decades of life and 2) minimize the decline in BMD after the age of 40 due to endocrine changes, aging, a decline in physical activity, and other factors. Because bone strength and resistance to fracture depend not only on the quantity of bone (estimated by BMD) but also bone geometry, methods are being developed that enable the assessment of cross-sectional geometry with existing DXA technology or with peripheral quantitative computed tomography (pQCT) or high-resolution magnetic resonance imaging (MRI). The microarchitecture of cancellous, or trabecular, bone (i.e., the lattice-work of bone inside vertebral bodies or ends of long bones) is important to the mechanical strength of the femoral neck, vertebral bodies, and other cancellous bone-rich regions. However, microarchitecture of cancellous bone can be assessed at present in humans only by bone biopsy, sophisticated MRI analyses, or the most advanced micro-CT devices not yet generally available. Additional valuable information can be gained from mechanical testing of bone samples from human cadavers and from animals subjected to various training protocols, and from histological and gene expression analyses from trained animals. Recent advances in protocols that enhance the osteogenic response to mechanical loading in animals have not yet been evaluated in humans, but are expected to stimulate new research in this area (116). The purpose of this position stand is to provide recommendations for the types of physical activities that are likely to promote bone health. The current state-of-knowledge regarding physical activity as it relates to 1) increasing peak bone mass, 2) minimizing age-related bone loss, and 3) preventing injurious falls and fractures will be discussed. ANIMAL STUDIES Various animal models have been utilized to study mechanical loading of the skeleton, but this section will focus mainly on the commonly used rat model. Multiple factors characterize the physical activities that are likely to influence properties of bone, including the type, intensity, duration, and frequency of the bone-loading activity. Studies of animals enable controlled manipulations of these factors to determine their relative contributions to the osteogenic response (i.e., bone formation). Type of loading Mechanical forces have osteogenic effects only if the stress to bone is unique, variable, and dynamic in nature. Static loading of bone (i.e., single, sustained force application) does not trigger the adaptive response that occurs with dynamic loading (11). Studies of rats have evaluated the osteogenic responses to several types of unique (i.e., not usual cage activity) exercise interventions, including running (treadmill and voluntary), swimming, jumping, standing, climbing, and resistance training. Results have been equivocal, demonstrating that mechanical stress can enhance (26,40,47,48,121,127,131) or compromise (8,26,92,132) bone mass, formation, and/or mechanical properties. In general, running and swimming of moderate intensity have been found to have positive effects on bone mass and material properties in the cortical and trabecular regions of the tibia and femur in growing and mature rats (8,26,47,121,127,131). However, decreases in bone mass, trabecular thinning, and structural properties have been observed in response to exercise that is very intense and/or excessive, particularly in growing animals (26,47,92,132). Activities that simulate resistance training in humans, including jumping up to a platform, voluntary tower climbing, and simulated “squat” exercises, have been found to have positive effects on both cortical and trabecular bone regions of the tibia and femur (91,92,126). Another experimental paradigm that has been used to evaluate the osteogenic effects of mechanical stress in animals is controlled in vivo external loading, including compression of the ulna and four-point bending of the tibia. This approach has an advantage over physical activity interventions in that it enables precise control and quantification of the mechanical loading forces. Studies of external loading strongly support favorable adaptations of bone to mechanical stress (116). For example, the four-point bending model was used in rats to demonstrate that the osteogenic response to loading is markedly enhanced when a given number of daily loading cycles are partitioned into multiple sessions separated by rest periods (116). It has not yet been determined whether such findings are relevant to humans. Intensity of loading The primary mechanical variables associated with load intensity include strain magnitude and strain rate. Strain is a measurement of the deformation of bone that results from an external load and is expressed as a ratio of the amount of deformation to the original length. It has long been recognized that strain magnitude is positively related to the osteogenic response, but accumulating evidence suggests that strain rate is also an important factor (11). Increasing strain rate, while holding loading frequency and peak strain magnitude constant, was found to be a positive determinant of changes in bone mass (11). High strain rates also increased endocortical bone formation rate in an in vivo impact-loading protocol (27,50). Such observations emphasize the need for further studies of the osteogenic effects of exercises that generate high strain magnitude and rate, such as jumping activities. Duration and frequency of loading The seminal work of Rubin and Lanyon (102) using external loading demonstrated that only a few loading cycles (e.g., 36 per day) of relatively high magnitude were necessary to optimize the bone formation response; increasing the number of loading cycles by 10-fold had no additional effect. Similarly, in a more physiologic model of loading in which rats jumped down from a height of 40 cm, as few as 5 jumps per day increased bone mass and strength of the tibia; increasing the number of jumps beyond 10 per day did not yield further benefit (118). It should be noted that, in these studies, the levels of strain likely exceeded those generated during typical human physical activities. The interactions between frequency (repetitions per day and sessions per week) and intensity of loading cycles with respect to the resulting osteogenic response in humans is not known. There is intriguing evidence from recent studies that applying a given number of loading cycles in multiple daily sessions is more osteogenic than applying the same number of cycles in a single daily session (116). Rat ulnas that were loaded 360 times per day in a single session (1×360) for 16 wk absorbed 94% more energy before failing than the contralateral unloaded ulnas. However, ulnas that received the same 360 daily loading cycles over 4 sessions (4×90) absorbed 165% more energy before failing than unloaded bones (116). These results suggest that bone cells lose sensitivity to mechanical stimulation after a certain number of loading cycles, and that recovery periods are needed to restore sensitivity to loading. It has been estimated that complete restoration of sensitivity to loading requires a recovery time of 8 h in rats, but recovery times as short as 0.5–1.0 h have been found to be more osteogenic than no recovery period (116). It will be important to determine in humans whether multiple, short daily exercise bouts are more osteogenic than a single, longer daily exercise session. Other considerations The ability of the skeleton to respond to mechanical loading can be either constrained or enabled by nutritional or endocrine factors. One example of this is calcium insufficiency, which diminishes the effectiveness of mechanical loading to increase bone mass (66). Another example is estrogen status. The independent effects of estrogen on bone metabolism are well described, but recent studies have determined that the adaptive response of bone cells to mechanical stress involves the estrogen receptor; blocking the estrogen receptor impairs the bone formation response to mechanical stress (133). This observation has led to the hypothesis that a down-regulation of estrogen receptors as a consequence of postmenopausal estrogen deficiency decreases the sensitivity of bone to mechanical loading. The mechanisms of mechanotransduction in bone (i.e., how mechanical forces are translated into metabolic signals) remain to be elucidated, and the discovery of key elements in the mechanistic pathways will likely reveal factors, potentially modifiable, that influence the osteogenic response to loading. As an example, it has been observed that prostaglandins and nitric oxide are produced by bone cells in response to mechanical loading, and that blocking their production impairs the bone formation response (16,115). The translation of such information generated from studies of animals and cultured bone cells will be critical in finding strategies to maximize the osteogenic effects of physical activity in humans. HUMAN STUDIES In humans, physical activity appears to play an important role in maximizing bone mass during childhood and the early adult years, maintaining bone mass through the fifth decade, attenuating bone loss with aging, and reducing falls and fractures in the elderly. The benefits of physical activity on bone health have typically been judged by measuring associations of physical activity level with bone mass and, in fewer studies, incidence of fractures, or by evaluating changes in bone mass that occur in response to a change in physical activity level or to a specific exercise training program. In evaluating the osteogenic effects of exercise training programs, the following principles should be noted: Specificity. Only skeletal sites exposed to a change in daily loading forces undergo adaptation. Overload. An adaptive response occurs only when the loading stimulus exceeds usual loading conditions; continued adaptation requires a progressively increasing overload. Reversibility. The benefits of exercise on bone may not persist if the exercise is markedly reduced. However, the rate at which bone is lost when an exercise program is discontinued, and whether this is different in young vs older individuals, is not well understood. The associations of physical activity and specific types of exercise with bone mass have been assessed in a variety of research paradigms. As reviewed previously (51,123), the majority of studies have been cross-sectional, comparing nonathletes with athletes who participate in a variety of sports, or comparing people who report being sedentary with those who report varying levels of physical activity. Because of the numerous confounding factors inherent to cross-sectional studies, these will be discussed only briefly. The response of bone to changes in physical activity and exercise training has also been assessed, including prospective studies (e.g., athletes through peak and training and controlled intervention studies in which physical activity is increased (e.g., exercise or (e.g., bed the most evidence that mechanical loading is essential to bone from studies of bed and which demonstrate that bone loss is rapid and profound when mechanical forces on the skeleton are markedly research is needed to the interactions of physical activity with and other factors, with respect to the influence on bone health. However, due to a of evidence to these issues will not be of physical activity in maximizing bone mass in children and adolescents A primary factor associated with risk for osteoporosis is the peak bone mass developed during childhood and the early adult suggest that trabecular bone loss as early as the decade, cortical bone increases or remains the fifth One study found that both cortical and trabecular bone mass continued to increase in young women well into the It has been observed that bone mass is in children who are physically active than in those who are active and in children who participate in activities that generate high impact forces (e.g., and than in those who in activities that impact forces (e.g., or are not (e.g., swimming) Recent studies have on jumping and other activities on the that high-intensity forces, greater gains in bone mass than to forces forces during jumping can 6–8 times and some generate forces that are times in ground-reaction forces during or running are times of the intervention studies of children were as of programs and between and months These studies found that children who in the experimental jumping and programs increased bone mass to a greater extent than children who in usual activities. One study that to other loading exercises found increases in bone mass of the and on this it is recommended that physical activity for children should include activities that generate relatively high ground-reaction forces, such as jumping, and running and, bone mineral accrual rate has been to occur at with of adult bone mineral a period of this time Thus, the period may a relatively short of time in which to maximize peak bone mass. studies that and athletes have BMD when with adolescents The effect is most in athletes who participate in sports that generate high-intensity or forces (e.g., gymnastics, lifting) and in athletes who participate in sports that generate loading forces. There have been few exercise intervention studies of all with significant changes in BMD were found in response to months of resistance training months of resistance training and with or months of and In significant increases in BMD in response to 3 of or months of resistance training The most between the studies that an effect of exercise and those that to do was the of the However, these studies a very small number of and must be There have been no studies that the effects of exercise training on the bone response, independent of changes in exercise volume or studies have to determine at in the period the skeleton is most to the benefits of physical activity or exercise training. One study determined the effect of months of and on bone mineral in and control were on status. increased in response to exercise in only Another study assessed the effect of months of on and BMD in and early and bone gains were observed in the early but not the when with A cross-sectional study evaluated BMD of both the and of with for of in BMD were in athletes and at but progressively in athletes at and with a at on these observations, bone appears to be most to mechanical stress during through to the that has been for peak bone mineral accrual the time of There remains a need for further research to the best and of exercise to augment bone accrual and the time during the period when loading is most The evidence to the same prescription noted previously for children (i.e., relatively high impact and activities, such as plyometrics, gymnastics, and resistance These activities appear to be most in bone mineral accrual when before or in the early Further, because measures of bone geometry may as important of bone strength that are independent of BMD and because it that geometric factors be particularly to mechanical stress during periods of it will be important to determine the influence of exercise on bone geometry in children and adolescents. of physical activity in young adults Because peak bone mass is to be by the of the decade, the early adult may be the for its cross-sectional studies of and athletes a variety of sports suggest that athletes have BMD when with nonathletes BMD to be in athletes who participate in sports that involve high-intensity loading forces, such as gymnastics, and and in athletes who participate in sports such as As noted inherent of cross-sectional studies include confounding variables such as and other factors. A of controlled studies of athletes have changes in bone mass through periods of training or in of level were greater than in and after 4 of Studies of and in from months to all significant increases in either or BMD of skeletal regions loaded by the specific of exercise performed during periods of training In for BMD increased during the and during the A number of intervention studies in from to 36 months have evaluated the effects of exercises that generate relatively high ground-reaction and/or forces (e.g., resistance training, on bone mass of previously sedentary women. The majority of these studies found significant increases in femoral and/or lumbar spine BMD In two of three studies of resistance training that to a significant effect on BMD, exercise intensity was only low to moderate (i.e., 60% or of 1-repetition Exercise intensity was high in the study (i.e., 5 10 4 but only the exercise was performed and this exercise may have for adaptation of the spine and femoral because it was performed in a position studies found an in BMD in response to relatively In one there was no change in femoral BMD but a in lumbar spine BMD after