The effects of various exercise protocols on bone tissue have received little attention compared to the attention given to the effects of exercise on muscle tissue. Although the losses in bone mineral content could limit the duration of future missions, the search for an optimal treatment has not been the primary concern.

Exercise is the principal countermeasure that is used in the spaceprograms (Grigoriev et al., 1992; Davis et al., 1996). Other possible countermeasures to counteract losses in bone mineral density, for instance dietary modifications (ingestion of supplementary calcium, phosphor and fluor) and hormonal supplements (such as anabolic steroids) have shown to be effective to some extent only (Lane et al., 1993; Davis et al., 1996). Pharmacological intervention, for instance the ingestion of biphosphonates, seemed to give positive results, but serious side-effects in patients receiving these drugs contra-indicated their further use (Anderson & Cohn, 1985; Minaire, 1989), although Russian researchers still recommend them (Grigoriev et al. 1992; Orlov et al., 1992). Another possible countermeasure to increase BMD is the application of low-frequency electrical fields around the bones of interest. Although this seems to be effective in increasing turkey tibial BMD, the effects on humans, including possible side-effects, are unknown (McLeod & Rubin, 1992).

The significant bone mineral losses during the Gemini IV and V flights in 1965 (see table 3) ignited the search for countermeasures in the United States flight program. For the Russian program, the stimulus was the determination of the major losses in calcaneal density of the Salyut I crew members, six years later. Countermeasures should be taken during, or even before, the engagement in spaceflight, since post-flight recovery is not self-evident (Tilton et al., 1980). Although recent research suggests that recovery is possible (LeBlanc & Schneider, 1991; Uebelhart, 1995), this would take several months (Chilibeck et al., 1995). Unlike muscle tissue, the rate of BMD recovery is slower than it’s rate of loss (LeBlanc et al., 1992). During this recovery period, muscular strength may well have reached it’s normal values long before bone tissue has recovered. This further increases the risks accompanying spaceflight-induced osteoporosis, because with the skeleton still affected, a person's own muscle strength can cause bone fractures (Bloomfield, 1997).

Until now, it has not been proven whether the degree of bone demineralization is altered by exercise (LeBlanc & Schneider, 1991; Drinkwater et al., 1995; Tipton et al., 1996). Consequently, researchers in spaceflight-induced physiological responses take great caution in prescribing any exercise protocol to counteract bone losses. In his elaborate review on the use of exercise as a countermeasure during prolonged spaceflight, Convertino extensively discusses cardiovascular effects, presents the findings in muscle function and merely mentions bone losses (Convertino, 1996). This approach reflects space-physiology related research pretty closely.

But indirect evidence is available to suggest that there is a relationship between exercise and a decreased degree of osteoporosis. A lot of recent research has been conducted in this area, mainly focusing on the beneficiary effects on the elderly. Although the differences in the target population are obvious, the basic ideas might be transferable. Despite the lack of indisputable evidence, the American College of Sport Medicine does recommend the incorporation of high intensity exercise in osteoporosis prevention programs (Drinkwater et al., 1995). And it is now generally accepted, that exercise is able to cause significant additions of both cortical and trabecular bone (Forwood & Burr, 1993).

It is important to recognize the exact nature of spaceflight-induced osteoporosis. It is commonly assumed that spaceflight-induced osteoporosis is synonymous to bone demineralization. Since the minerals give bone it’s strength, the use of this synonym does indicate the seriousness of the problem. However, it also implies that no other mechanism is involved, which is a faulty assumption.

During spaceflight, the molar changes in urinary calcium output are followed by molar changes in hydroxyproline and hydroxylysine (Tipton & Hargens, 1996). Hydroxyproline is a metabolite of the amino acid proline, which is a valid indicator of actual bone resorption. This finding implies that a loss of collagen has taken place as well as a loss of minerals (Zernicke et al., 1990; Lane et al., 1993). Thus, bone loss in space is a result of bone atrophy and not of mere bone demineralization. The recognition of this fact implies that spaceflight induced osteoporosis should be treated as true osteoporosis, i.e. as a complex mechanism, with interacting mechanical, humoral and immunal factors. Exercise is able to act on every aspect of this complex mechanism (Bosch, 1993; Taylor, 1993; Chilibeck et al., 1995).

Spaceflight-induced osteoporosis is believed to be caused by the relative disuse of the skeleton (Mack et al., 1967; Tipton & Hargens, 1996). In microgravity, the mechanical problem of not being able to place enough force on the bones to create sufficient osteogenic activity is worsened by kinematic changes in walking under microgravity conditions. In such conditions, one makes shorter steps and one tends to lean forward, thereby landing on the forefoot. This could mean reduced shock loading and decreased force placed on, among other tissues, the bones (Davis & Cavanagh, 1993; Davis et al., 1996; Wickman & Luna, 1996). In addition, when gravity drops below 40 per cent of the earth gravitational level, human locomotion changes in another way. The ground reaction force seems to be lessened more severely than the tangential component of walking (i.e. the circular movement of the centre of gravity around the forefoot). This results in a lack of traction under the forefoot and an even lower mechanical loading of the bones (Davis & Cavanagh, 1993).

It is not indisputably clear whether the microstructure of bones is affected due to spaceflight. In rats, the trabeculae, the internal structural network of bone, have become thinner after seven days of spaceflight (Vico & Alexandre, 1992). Taking bone biopsies is a very painful procedure, and have not been reported in human research regarding spaceflight. The only reported spaceflight-related research in this area has been performed on the three cosmonauts of Salyut I. They were tragically killed during their return to earth after a sudden decompression of their cabin. The microscopic and crystalline structure of their bone tissue did not differ from three earth based control subjects, who were of the same age and died of an acute trauma, unrelated to any bone disease (Rambaut & Goode, 1985; Arnaud & Morey-Holton, 1990).

These findings cannot be generalised because of the small number of subjects and the flaws in the experimental design. For example, it is possible that the cosmonauts had a higher pre-flight BMD compared to the controls. Therefore, until more bone biopsies from spacetravellers are obtained the short-term and long-term structural and compositional changes of bone remain speculative (Zernicke et al., 1990).

The underlying mechanism of bone demineralization has been elucidated, at least in mice. In the past, two mechanisms have been proposed that could account for the BMD losses in test animals, namely the lack of gravitational loading or hormonal changes, induced by the psychological stress of an involuntary spaceflight (Wronski & Morey, 1983). The first mechanism has been shown to have enough impact on bone to induce the observed changes. Microgravity, and the subsequenting relative unloading of the bones results in an increased bone resorption and a decreased bone formation (Durnova et al., 1991; Loon et al., 1995). The exact sequence of these two effects in humans is still not fully understood (Uebelhart et al., 1995).

Within a society with a growing percentage of elderly people, research in age-related diseases, like osteoporosis, is also increased. This has led to numerous recent publications focusing on this topic, many of which study the effects of various exercises on BMD losses (Gutin & Kasper, 1992; Swezey, 1996). The thought that exercise could serve as a medicine against osteoporosis gained popularity in the late 1980s. Before that time hormonal treatments were the predominant prescription (Minaire, 1989). The current belief is that exercise is an economic alternative to prescription drugs with a relatively low risk of serious side effects (Swezey, 1996).

Longitudinal research has demonstrated that people with an active lifestyle have greater BMD than people with a dominantly sedentary lifestyle (Chilibeck et al., 1995). The performance of physical activity during youth leads to a higher peak bone mass and has the ability to lessen bone mineral losses or even increase BMD thereafter (Botden & Kemper, 1996). The possibility of increasing BMD after skeletal maturity has been reached, has also been questioned. Although increments of one to three percent have been found in longitudinal studies, it is questionable whether these changes are within the range of measurement precision (Forwood & Burr, 1993). However, when physical exercise solely conserves BMD while losses would be expected, it is still a very useful medicine.

Weight-bearing activities are generally believed to be of great importance in preventing bone loss (Drinkwater et al., 1995). Such activities, as well as bodyweight itself, have a greater effect on BMD than calcium intake, at least when calcium intake is not too low (Welten et al., 1994). The importance of weight-bearing activities on BMD is also indicated by the significant relationship between bodyweight and spinal BMD (Stupakov et al., 1990) and in the finding that lumbar BMD of high lesion spinal cord-injured patients remains within normal range. Apparently, the continued load-bearing on the spine while seated in a wheelchair is enough stimulus to maintain BMD, despite the paralysis of adjacent muscles (Bloomfield, 1997). These results emphasize the importance of bone loading in preserving or enhancing bone mass.

The importance of weight-bearing activities in enhancing BMD is also seen in several studies that included non-weight-bearing exercises. In bed rest studies, supine leg exercise has been shown not to have any effects on lumbar BMD (Arnaud & Morey-Holton, 1990), while three hours of standing each day did slow bone losses (Chilibeck et al., 1995). Similarly, moderate walking, which has a small weight-bearing component, has been shown to be insufficient to provide much protection against bone loss in the elderly (Gutin & Kasper, 1992).

The effects of exercise appear to have a predominant site-specific effect. Exercises have been shown to have the greatest effect on the bones that are directly subjected to the application of force (Gutin & Kasper, 1992; Forwood & Burr, 1993). This means that forearm BMD is mainly under the influence of arm exercises, while axial loading has the greatest effect on BMD of the spine (Groothausen et al., 1997). Consequently, to counteract BMD losses in the lower extremities, these limbs should be included in exercise protocols.

Although mere loading appears to be a sufficient stimulus to influence bone mass, even better effects have been found with strenuous exercise of short duration. The high strains on bones that accompany this kind of exercise, locally increase bone density (Schoutens et al., 1989; Swezey, 1996). Sometimes the increase in BMD is attended with bone hypertrophy (Gutin & Kasper, 1992). The performance of activities which induce high peak strains on the spine, such as ballet, gymnastics, volleyball and playing squash is a significant positive predictor for lumbar peak bone mass (Groothausen et al., 1997).

To counteract the degeneration of the bone, stresses should be imposed on it since mechanical stress is a functional stimulus to increase osteoblast-activity. This assumption is based on a statement by Julius Wolff in 1868. Known as "Wolff’s law", it is still accepted as an accurate description of the functional relationship between bones and external forces working upon them (Busby, 1968; Duncan & Turner, 1995).

"Every change in the form and function of bones, or in their function alone, is followed by certain definite changes in their internal architecture, and equally definite changes in their external conformation." (from Busby, 1968)

Busby transferred these common descriptions to the practice of space-physiology as follows:

"By means of formation and reabsorption processes, bones are remodelled in response to the functional demands which muscular pull, other mechanical stresses, and presumably gravitational forces place on them."

The creation of artificial gravitational forces inside a spacecraft has often been considered. However, creating artificial gravitational forces during spaceflight is energetically very costly since it would require a continuously rotating spacecraft. Besides, people would develop motion sickness due to the coriolis effect from head motions inside such a spacecraft (Convertino, 1996). Consequently, other equivalents of weight-bearing activities should be used in enhancing BMD in a weightless environment. Thus, the objective of exercise protocols during space-missions must be to generate a muscular pull and/or mechanical stresses great enough to generate a proper remodelling response at the appropriate sites, i.e. the skeletal sites which suffer the most BMD loss during spaceflight.

The mechanical forces that work on bone (generated by either muscular pull or external mechanical stresses) are quantified as stresses and strains. Stress is defined as the load per unit area developed on a bone surface due to externally applied loads. Strain is the deformation that occurs in response to this loading, where one microstrain equals one micrometer of deformation per meter of length (Chilibeck et al., 1995; Duncan & Turner, 1995). Strain magnitude is believed to determine the response of the bone tissue under strain. This is the basis of the "mechanostat" hypothesis put forth by Frost (Frost, 1988; Duncan & Turner, 1995).

The mechanostat theory is based on the idea that bone remodels itself to the strength that is needed to cope with the demands placed upon it. Whenever a certain minimum effective strain is surpassed, the bone under strain adapts to this changed demand. Exercise, especially when it involves sudden impact forces (i.e. high peak forces produced at a high rate), causes the development of stronger bones (Frost, 1988; Duncan & Turner, 1995).

The basic thought in Wolff’s law and the mechanostat theory is functionality. In accordance with the near disappearance of gravity during spaceflight and the consequent relative unloading of the skeleton, bone density is lowered. The objective of executing exercise protocols in-flight is to increase the functional demands on bone and consequently increasing it’s density.

The evidence clearly suggests that bones under strain increase their BMD. How mechanical stresses are transducted towards bone remodeling is not clear. A lot of research is done in this area, and several mechanisms have been proposed, which is in striking contrast with the meagre theories that underlie the effects of exercise on muscle tissue. It is surprising (however maybe even logical) that an effect which is difficult to prove irrefutably is substructured with several possible mechanisms on the cellular level, while a well-known effect which existence has been proven extensively has very little substructing evidence.

Five mechanisms have been proposed through which mechanical stimuli may result in increased bone formation (Chilibeck et al., 1995). Decreased bone resorption may also be an effect of one (or a combination) of these stimuli (Duncan & Turner, 1995). The five mechanisms are: prostaglandin release, hormonal alterations, formation in response to microdamage, piezoelectrically-induced potential changes and increased bone blood flow.

The first two mechanisms, prostaglandin release and hormonal alterations, are in favour of a systemic bone enhancement response, which means that bone formation is stimulated all over the body. Animal studies have shown that prostaglandin is released after exercise, and dinoprostone, or prostaglandin E₂, increases bone strength. Exercise also leads to increased levels of several bone enhancing hormones, like calcitriol and testosterone (Chilibeck et al., 1995).

The final three mechanisms, namely formation in response to microdamage, piezoelectrically-induced potential changes and increased bone blood flow, imply a pure local effect of mechanical stimuli. Increased bone formation after microdamage of bone structures is also known as woven bone formation (Forwood & Burr, 1993). This kind of damage is present after the application of high forces on a specific bone, e.g. after strenuous exercise. Piezoelectricity is the generation of an electric potential after deformation. Crystals, like quartz, lack a centre of symmetry and this causes, once the crystal is deformed, a separation of opposite charges. The compression side becomes negatively charged; the tensile side positively. New bone is formed on the compression side, maybe because the negative charge attracts positively charged calcium ions. Bone may be a piezoelectric substance, since collagen and hydroxyapatite, the chief bone mineral, exist in a crystalline state. On the other hand, piezoelectricity is more notable in dry tissue, which bone is not (Schultheis, 1991). Compression also leads to changes in bone blood flow, which is the basis of the final possible mechanism. Once blood flow is increased in places that are usually less perfused, these areas experience a better nutrition, which facilitates the osteogenic processes.

It is unknown what mechanism has the greatest effect on BMD. Since mechanical stimuli have a predominantly local effect on BMD, according to the law of site specificity, one of the mechanisms that has a local effect should be dominant. But the systemic factors cannot be ruled out. In a study by Snow-Harter et al. (1992), besides several expected correlations between muscle strength and BMD of their bones of insertion, a correlation was found between biceps strength and trochanter major BMD. From these results, it can be concluded that the relationship between strength and BMD is more complex than a simple site-specific model.

Schultheis (1991) expects that a combination of piezoelectrically-induced potential changes and increased bone blood flow is physiologically the most important. Compression also leads to flows in interstitial fluid, and the ions that are dissolved in it. These flows are called "streaming potentials" and result in a negative charge on the compression side of the bone, which enhances bone formation. Until the exact underlying mechanism has been elucidated, the formation of exercise protocols to increase BMD can only be based upon the results of longitudinal studies and upon the findings that high strains stimulate bone formation.

Although the mechanostat theory is based on the assumption that strain magnitude determines the bone enhancing effects, this is not definite. Sudden impact loading usually implies both a high maximal ground reaction force (GRF-max) and a high rate of change in force while producing this maximum (dF/dt). In the bone, both the strain magnitude and the rate of change in strain are high. It is still uncertain which of these aspects is most efficient in enhancing bone mineral density (Cavanagh et al., 1992; Whalen et al., 1993; Duncan & Turner, 1995). Maybe it is not necessary to focus on producing a high GRF-max while exercising in space. It could be sufficient to conduct exercises that incorporate a sufficiently high maximum dF/dt, which is easier to achieve in a weightless environment (Davis & Cavanagh, 1993). This possibility should be studied more precisely, before efforts are made to create high GRF-maxs under weightless conditions.

The conclusion can be drawn that bone is a tissue with dynamic properties (Zernicke et al., 1990; Snow-Harter & Marcus, 1991). The skeletal system homeostasis of bone formation and resorption is rapidly altered when the functional loading of the skeleton is changed (Minaire, 1989; Chilibeck et al., 1995). This quick adaptation response also gives hope to the possibilities of counteracting these changes. Although continuous loading is probably impossible to achieve during spaceflight, daily intermittent exercises that involve high impact loading might lead to a functional change within the skeleton.

Specifications of effective, BMD enhancing, exercise in terms of site of application, type, intensity, frequency and duration are still unknown (Schoutens et al., 1989; Groothausen et al., 1997). Consequently, the prescription of exercise during long-term spaceflights is based on theory more than on experience, and in-flight exercise protocols cannot be fully accepted until their success has been validated during actual spaceflight. But, based on the findings that exercise is able to cause significant additions in BMD and using generally accepted theories, certain assumptions can be made.

In the establishment of the contents of exercise protocols aimed at minimizing BMD losses, Wolff’s law and the mechanostat theory should be borne in mind. These theories form the basis of how exercise, manifested as mechanical strain, can cause physiological changes in bone tissue. Based on these theories, general features of exercise protocols can be determined, after which an optimal protocol can be designed for each individual engaging in long-term spaceflight, since individual differences will always exist.

Operational demands enforce an extra restriction on possible in-flight exercises. The production of high strains on the skeleton must not be accompanied by vibrations of the spacecraft, as these can cause problems to the spacecraft’s dynamics and to on-board experiments (Cavanagh et al., 1992). The NASA is developing a platform on which vibrating exercises can be performed without causing any disturbances outside the platform (Bachl et al., 1993).

Strain seems to be the most important issue in enhancing bone mass. As long as the mechanical impact on bones is severe enough, the bone under strain will become stronger in order to cope with such impacts. Whether it is the sudden change in strain or the magnitude of strain that results in these changes is still uncertain. It would be advisable to determine this, before time and money is spend to produce high peak strains in-flight. Until this fundamental question has been resolved, the prescription of exercises shall continue to remain in general terms.

All bones are affected by the decreased loading in the absence of gravity. Therefore, all bones should be subjected to an in-flight exercise protocol. Although trabecular bone has been shown to be affected the most by unloading, this type of bone does not need extra attention. Exercise has concomitantly been shown to be more effective in enhancing trabecular bone than cortical bone, probably because the network-structure of trabecular bone allows more surface area to be under the influence of osteoblast and osteoclast activity (Gutin & Kasper, 1992).

The challenge is to determine exercises, that produce strains that have been proven to increase BMD. An example of such an exercise, is tennis ball squeezing. One year of tennis ball squeezing for 30 seconds a day did increase radial BMD significantly in 77 elderly women (Chilibeck et al., 1995). This result shows that an easy task, which is not dependent on gravitational forces, can have significant results. Tennis ball squeezing is easy to perform during spaceflight, and it involves cheap and light-weight equipment, thus making it the perfect countermeasure against spaceflight related problems. But, according to the law of site specificity, this specific exercise only acts on forearm BMD. Thus, other exercises should be sought to have the same impact on other sites (Schoutens et al., 1989).

High-impact resistance exercise seems to be able to elicit a bone-enhancing response by placing strains of a sufficient magnitude on the skeleton. This kind of exercise can be performed under microgravity by squeezing tennis balls and with standard resistance exercise equipment, as long as the resistance is delivered by spring-like equipment and not by weights. These exercises elicit the sudden impact shocks that are believed to increase BMD (Davis et al., 1996; Swezey, 1996). These exercises should include all skeletal sites of the legs, spine and arms, although emphasis should be placed on the weight-bearing bones. Whole body exercise should also be included in an in-flight exercise protocol to increase BMD. It is believed that stabilising contractions of the trunk musculature could contribute more to spinal BMD than specific exercises that involve only the muscle attached to the spine (Chilibeck et al., 1995).