#

**Controls**

**\* p < 0,05 vs control # p < 0,05 vs high**

**\* \* \* \***

Fig. 3. The picture depicts the analysis of bone mineral density (BMD) in high and low level paraplegics and controls. A statistically significant reduction in total BMD (p<0.001) and lower limbs BMD in body composition compared to able-bodied males was observed. On the contrary, upper limbs BMD was higher in low paraplegics and controls, an unexpected

**Lower Limbs BMD Upper Limbs-BMD Total BMD**

finding explained in the paper of Dionyssiotis et al., 2008b. Diagram modified and

translated from Dionyssiotis, 2008a.

**0**

**0.2**

**0.4**

**0.6**

**mean value in grcm-2**

**0.8**

**1**

**1.2**

**1.4**

most possible reason to give falsely higher values of BMD (Dauty et al., 2000).

the proximal tibia (-70%), respectively.

The neurological level of the lesion i.e. the extent of impairment of motor and sensory function is important, because tetraplegics are more likely to lose more bone mass throughout the skeleton than paraplegics (Tsuzuku et al., 1999). In paraplegics legs' BMC was reduced vs. controls, independently of the neurological level of injury and negatively correlated with the duration of paralysis in total paraplegic group, but after investigation according to the neurological level of injury this correlation was due to the strong correlation of high paraplegics' legs BMC with the duration of paralysis, meaning that the neurological level of injury determines the extent of bone loss (Dionyssiotis et al., 2009). The similar severity of demineralization in the sublesional area was shown between paraplegics and tetraplegics, and the extent of the bone loss may be variable (Demirel et al., 1998; Tsuzuku et al., 1999; Dauty et al., 2000).

The duration of paralysis has an inverse relationship with leg percentage-matched BMD and trunk percentage-matched BMD (Clasey et al., 2004). In addition in complete paraplegics, with high (thoracic 4-7) and low (thoracic 8-12) neurological level of injury, upper limbs FM and lower limbs BMD were correlated with the duration of paralysis in total paraplegic group but after investigation according the neurological level of injury this correlation was due to the strong correlation of high paraplegics' lower limbs BMD with the duration of paralysis. The explanation of this strong correlation could possibly lie on higher incidence of standing in the group of low paraplegics and direct effect of loading lower limbs while standing and walking with orthotic equipment. Moreover, the association of the duration of paralysis with parameters below and above the neurological level of injury (upper limbs FM) raises the question of the existence of a hormonal mechanism as an influential regulator in paraplegics' body composition (Dionyssiotis, 2008a; Dionyssiotis et al., 2008b; 2009).

Actually, little is known regarding the nature and time frame of the influence of complete SCI on human skeletal muscle because published data are coming from cross-sectional studies, where different groups with few subjects have been examined at different times, usually in the chronic phase of paralysis. Disuse was thought to be the mechanism responsible for the skeletal muscle atrophy in paraplegics, but muscle fibres following SCI begin to change their functional properties early post injury. Muscle fiber cross-sectional area (CSA) has been suggested to decline from 1 to 17 months after injury and thereafter to reach its nadir. Conversion to type II fibers has been suggested to occur between 4 months and 2 years after injury, resulting in even slow-twitch muscle becoming predominantly fast twitch thereafter (Castro et al., 1999). Metabolic enzymes levels in skeletal muscle might be expected to be reduced after SCI because of inactivation. In support of this contention, succinic dehydrogenase (SDH) activity, a marker of aerobic-oxidative capacity, has been reported to be 47–68% below control values in fibers of tibialis anterior muscle years after injury in support of this contention (Scelsi, 2001).

The muscle atrophy in SCI is of central type and depends on the disuse and loss of upper connections of the lower motor neuron, sometimes associated to the loss of anterior horn cells and transinaptic degeneration. The last alteration may be responsible for the denervation changes seen in early stages post SCI. In the later stages (10-17 months post SCI) diffuse muscle atrophy with reduction of the muscle fascicle dimension is associated to fat infiltration and endomysial fibrosis. In all stages post SCI, almost all patients showed myopathic changes, as internal nuclei, fibre degeneration and cytoplasmic vacuolation due to lipid accumulation (Scelsi, 2001)

Body Composition in Disabilities of Central Nervous System 85

injury in most patients with SCI to develop autonomic dysreflexia. With SCIs above the level of T6, there is reduced SNS outflow and supraspinal control to the splanchnic outflow and the lower-extremity blood vessels while serum leptin levels in men with SCI correlated not only with BMI but also with the neurologic deficit. This finding supports the notion that decentralization of sympathetic nervous activity relieves its inhibitory tone on leptin secretion, because subjects with tetraplegia have a more severe deficit of sympathetic

No significant difference between ambulatory multiple sclerosis (MS) patients and non MS controls in body composition was found despite lower physical activity in ambulatory MS patients (Lambert et al., 2002). In MS subjects there was no significant relation between any of the body composition measures and the level of disability as measured by the Expanded Disability Status Scale (EDSS). Others found no difference in body fat percent between ambulatory MS patients (Formica et al., 1997) and lower physical activity in ambulatory MS patients vs. controls (Ng & Kent-Braun, 1997). A possible explanation for the similar body composition may be lower energy intake in MS individuals who are ambulatory and greater energy cost of physical activity (walking) in MS than it is with non MS controls (Lambert et

A significant inverse relation between free fat mass (FFM) and EDSS score when ambulatory and non ambulatory MS subjects were combined was found (Formica et al., 1997). On the contrary others without including non ambulatory subjects did not find a significant inverse relation between FFM percent and EDSS score (Lambert et al., 2002). It would seem apparent that ambulatory patients with MS and controls would strengthen the inverse

The finding of no relation between EDSS score and body fat percent (Lambert et al., 2002) fits well with studies which found no significant relation between the level of physical activity, and the level of disability in individuals with MS (Ng & Kent-Braun, 1997) because MS would likely have a much greater effect on physical activity than on energy intake. According to these findings it appears that the level of disability of ambulatory individuals with MS does not predict body composition. This suggests that a significant level of disability does not force these individuals to be physically inactive and does not result in a greater body fat content. There are many detrimental manifestations of excess body fat, such as hyperlipidemia, insulin resistance, and type II diabetes (Lambert et al., 2002). The largest component of FFM is muscle mass (Lohman, 1986). If muscle mass is lower in individuals with MS than in controls, it may also contribute to the impaired ability to ambulate and perform other activities of daily living. Muscle fiber size from biopsy specimens of the tibialis anterior were 26% smaller than specimens from control subjects (Kent-Braun et al., 1997). Thus, at least for this small muscle, muscle mass was lower in MS. This relationship may not hold for other muscle groups or for whole-body muscle mass (Lambert et al., 2002). Another reason for skeletal muscle alterations is glucocorticoid usage. The prolonged duration of glucocorticoid causes catabolism of skeletal muscle. Decreased amino acid transport into muscle and increased glutamine synthesis activity with resultant muscle atrophy are some of the concomitant effects of glucocorticoid use on skeletal muscle.

nervous activity (Wang et al., 2005).

relation between FFM and EDSS score.

**3.2 Multiple sclerosis** 

al., 2002).

It is evident that other co-factors as spasticity and microvascular damage, contribute to the induction of the marked morphological and enzyme histochemical changes seen in the paralyzed skeletal muscle (Scelsi, 2001). Small fibers, predominantly fast-twitch muscle, and low mitochondrial content have been reported years after injury in cross-sectional studies. These data have been interpreted to suggest that human skeletal muscle shows plasticity (Castro et al., 1999).

On the contrary, force loss during repetitive contractions evoked by surface electrical stimulation (ES) of skeletal muscle in humans does not appear to be altered within a few months of injury (Shields, 1995) but it is greater a year or more after SCI (Hillegass & Dudley, unpublished observations). The greater fatigue, when evident, was partially attributed to lower metabolic enzyme levels (Scelsi, 2001).

Muscular loading of the bones has been thought to play a role in the maintenance of bone density (de Bruin et al., 1999; Dionyssiotis et al., 2011d). However, the ability to stand or ambulate itself does not improve BMD or prevent osteoporosis after SCI.

Controversial results have also been reported regarding the effect of spasticity on BMD in SCI paraplegics. A cross-sectional study of 41 SCI paraplegics reported less reduction of BMD in the spastic paraplegics SCI patients compared to the flaccid paraplegic SCI patients (Demirel et al., 1998). Others reported that spasticity may be protective against bone loss in SCI patients, however, without any preserving effect in the tibia (Dionyssiotis et al., 2011a; Eser et al., 2005). A possible explanation for that could lie in the fact paraplegics to be above thoracic (T)12 level with various degrees of spasticity according to the Ashworth scale. In addition, muscle spasms affecting the lower leg would mainly be extension spasms resulting in plantar flexion thus creating little resistance to the contracting muscles. Furthermore, the measuring sites of the tibia did not include any muscle insertions of either the knee or the ankle extensor muscles (Dionyssiotis et al., 2011a, 2011d). Other investigators also have not been able to establish a correlation between BMD and muscle spasticity (Lofvenmark et al., 2009).

The hormone leptin is secreted by fat cells and helps regulate body weight and energy consumption (Fruhbeck et al., 1998). The percentage of fat in people is positively correlated with the amount of leptin in the circulation (Maffei et al., 1995). In SCI, when compared with healthy subjects, higher levels of leptin have been found, possibly due to greater fat tissue storage (Bauman et al., 1996). Leptin activates the sympathetic nervous system (SNS) through a central administration. The disruption of the sympathetic nervous system i.e. in tetraplegia and high level paraplegia may modify the secretion and activity of the leptin, because the sympathetic preganglionic neurons become atrophic in these subgroups (Elias et al., 1998; Correia et al., 2001) leading to disturbed irritation from leptin below the neurological level of injury. In addition, extensive obesity is known to reduce lipolytic sensitivity (Haque et al., 1999; Horowitz et al., 1999, 2000).

In high level spinal cord injuries there is a disorder of the autonomic nervous system and combined to the fact that the hormone leptin activates the sympathetic nervous system through central control it could be suggested that "the closure of paths" of the central nervous system disrupts the effect of leptin and possibly increases the risk of obesity in SCI subjects with high-level injury (Krassioukov et al., 1999; Jeon et al., 2003). However, after separation of SCI subjects into those with an injury above or below Thoracic (T) 6, leptin levels were significantly higher in the former group. T6 appears to be the lowest level of injury in most patients with SCI to develop autonomic dysreflexia. With SCIs above the level of T6, there is reduced SNS outflow and supraspinal control to the splanchnic outflow and the lower-extremity blood vessels while serum leptin levels in men with SCI correlated not only with BMI but also with the neurologic deficit. This finding supports the notion that decentralization of sympathetic nervous activity relieves its inhibitory tone on leptin secretion, because subjects with tetraplegia have a more severe deficit of sympathetic nervous activity (Wang et al., 2005).

#### **3.2 Multiple sclerosis**

84 Dual Energy X-Ray Absorptiometry

It is evident that other co-factors as spasticity and microvascular damage, contribute to the induction of the marked morphological and enzyme histochemical changes seen in the paralyzed skeletal muscle (Scelsi, 2001). Small fibers, predominantly fast-twitch muscle, and low mitochondrial content have been reported years after injury in cross-sectional studies. These data have been interpreted to suggest that human skeletal muscle shows plasticity

On the contrary, force loss during repetitive contractions evoked by surface electrical stimulation (ES) of skeletal muscle in humans does not appear to be altered within a few months of injury (Shields, 1995) but it is greater a year or more after SCI (Hillegass & Dudley, unpublished observations). The greater fatigue, when evident, was partially

Muscular loading of the bones has been thought to play a role in the maintenance of bone density (de Bruin et al., 1999; Dionyssiotis et al., 2011d). However, the ability to stand or

Controversial results have also been reported regarding the effect of spasticity on BMD in SCI paraplegics. A cross-sectional study of 41 SCI paraplegics reported less reduction of BMD in the spastic paraplegics SCI patients compared to the flaccid paraplegic SCI patients (Demirel et al., 1998). Others reported that spasticity may be protective against bone loss in SCI patients, however, without any preserving effect in the tibia (Dionyssiotis et al., 2011a; Eser et al., 2005). A possible explanation for that could lie in the fact paraplegics to be above thoracic (T)12 level with various degrees of spasticity according to the Ashworth scale. In addition, muscle spasms affecting the lower leg would mainly be extension spasms resulting in plantar flexion thus creating little resistance to the contracting muscles. Furthermore, the measuring sites of the tibia did not include any muscle insertions of either the knee or the ankle extensor muscles (Dionyssiotis et al., 2011a, 2011d). Other investigators also have not been able to establish a

The hormone leptin is secreted by fat cells and helps regulate body weight and energy consumption (Fruhbeck et al., 1998). The percentage of fat in people is positively correlated with the amount of leptin in the circulation (Maffei et al., 1995). In SCI, when compared with healthy subjects, higher levels of leptin have been found, possibly due to greater fat tissue storage (Bauman et al., 1996). Leptin activates the sympathetic nervous system (SNS) through a central administration. The disruption of the sympathetic nervous system i.e. in tetraplegia and high level paraplegia may modify the secretion and activity of the leptin, because the sympathetic preganglionic neurons become atrophic in these subgroups (Elias et al., 1998; Correia et al., 2001) leading to disturbed irritation from leptin below the neurological level of injury. In addition, extensive obesity is known to reduce lipolytic

In high level spinal cord injuries there is a disorder of the autonomic nervous system and combined to the fact that the hormone leptin activates the sympathetic nervous system through central control it could be suggested that "the closure of paths" of the central nervous system disrupts the effect of leptin and possibly increases the risk of obesity in SCI subjects with high-level injury (Krassioukov et al., 1999; Jeon et al., 2003). However, after separation of SCI subjects into those with an injury above or below Thoracic (T) 6, leptin levels were significantly higher in the former group. T6 appears to be the lowest level of

attributed to lower metabolic enzyme levels (Scelsi, 2001).

ambulate itself does not improve BMD or prevent osteoporosis after SCI.

correlation between BMD and muscle spasticity (Lofvenmark et al., 2009).

sensitivity (Haque et al., 1999; Horowitz et al., 1999, 2000).

(Castro et al., 1999).

No significant difference between ambulatory multiple sclerosis (MS) patients and non MS controls in body composition was found despite lower physical activity in ambulatory MS patients (Lambert et al., 2002). In MS subjects there was no significant relation between any of the body composition measures and the level of disability as measured by the Expanded Disability Status Scale (EDSS). Others found no difference in body fat percent between ambulatory MS patients (Formica et al., 1997) and lower physical activity in ambulatory MS patients vs. controls (Ng & Kent-Braun, 1997). A possible explanation for the similar body composition may be lower energy intake in MS individuals who are ambulatory and greater energy cost of physical activity (walking) in MS than it is with non MS controls (Lambert et al., 2002).

A significant inverse relation between free fat mass (FFM) and EDSS score when ambulatory and non ambulatory MS subjects were combined was found (Formica et al., 1997). On the contrary others without including non ambulatory subjects did not find a significant inverse relation between FFM percent and EDSS score (Lambert et al., 2002). It would seem apparent that ambulatory patients with MS and controls would strengthen the inverse relation between FFM and EDSS score.

The finding of no relation between EDSS score and body fat percent (Lambert et al., 2002) fits well with studies which found no significant relation between the level of physical activity, and the level of disability in individuals with MS (Ng & Kent-Braun, 1997) because MS would likely have a much greater effect on physical activity than on energy intake. According to these findings it appears that the level of disability of ambulatory individuals with MS does not predict body composition. This suggests that a significant level of disability does not force these individuals to be physically inactive and does not result in a greater body fat content. There are many detrimental manifestations of excess body fat, such as hyperlipidemia, insulin resistance, and type II diabetes (Lambert et al., 2002). The largest component of FFM is muscle mass (Lohman, 1986). If muscle mass is lower in individuals with MS than in controls, it may also contribute to the impaired ability to ambulate and perform other activities of daily living. Muscle fiber size from biopsy specimens of the tibialis anterior were 26% smaller than specimens from control subjects (Kent-Braun et al., 1997). Thus, at least for this small muscle, muscle mass was lower in MS. This relationship may not hold for other muscle groups or for whole-body muscle mass (Lambert et al., 2002).

Another reason for skeletal muscle alterations is glucocorticoid usage. The prolonged duration of glucocorticoid causes catabolism of skeletal muscle. Decreased amino acid transport into muscle and increased glutamine synthesis activity with resultant muscle atrophy are some of the concomitant effects of glucocorticoid use on skeletal muscle.

Body Composition in Disabilities of Central Nervous System 87

percentile for age, even if most of the CP children had a low height and weight for age. In female subjects anthropometric measurements were highly correlated with measures of body fatness. Measuring fat by 18O dilution a hydration factor of 0.73 was assumed for FFM. A possible increase in the hydration factor would diminish measured FFM meaning that body fat appears increased. Moreover muscle spasms and spasticity in CP subjects deplete body glycogen. If glycogen is reduced the intracellular water would be reduced and the ratio extracellular water/total body water would increase. The same could result with a loss

Other important issues according alterations of body composition are the completeness of lesions (an absence of sensory or motor function below the neurological level, including the lowest sacral segment), because body composition seems to be worst than subjects with incomplete lesions (partial preservation of motor and/or sensory function below the neurological level, including the lowest sacral segment) (Sabo et al., 1991; Demirel et al., 1998; Garland et al., 1992) and aging which contributes to major alterations of body

In disabled subjects the most important issue according to body composition is how to promote optimal body weight to reduce risk of diseases such as coronary heart disease, noninsulin dependent diabetes mellitus, lipid abnormalities and fractures because of bone loss. Dietary changes, individualized physical activity programs and medication should be taken in mind in therapy when we deal with this subgroup of subjects. However, selfmanagement of dietary changes to improve weight control and disease should be the case, which means they need to follow diets with lower energy intake and at the same time to eat

We need to take in mind that healthy BMI values often underestimate body fat and may mask the adiposity and spasticity did not defend skeletal muscle mass and bone, supporting the concept that in neurologic disabilities the myopathic muscle could not recognize correctly the stimulation because of the neurogenic injury. Moreover, disabled subjects mostly transfer much of the weight-bearing demands of daily activities to their upper extremities reducing the weight-bearing of the affected paralyzed muscles triggering a cycle of added muscle atrophy which interacts with the continuous catabolic action caused by the neurogenic factor. Finally, an irreversible (once established) decline in bone mineral density, bone mineral content as well as geometric characteristics of bone is expected and the duration of lesion-injury is positively correlated with the degree of

Further research about body composition is needed in all physical disabilities and more longitudinal studies to quantitate and monitor body composition changes and to modify our therapeutic interventions. However, prevention rather than treatment may have the greatest potential to alleviate these major complications. Therapies should focus on how to perform weight bearing, standing or therapeutically walking activities early in the rehabilitation

of body cell mass or an increase in the hydration factor (Bandini et al., 1991).

regularly foods rich in nutrients (Groah et al., 2009).

program to gain benefits according to muscles and bones.

**4. Conclusions** 

composition.

bone loss.

Endogenous glucocorticoid excess also produces generalized osteoporosis, most prevalent in trabecular-rich skeletal regions (Formica et al., 1997).

Beside corticosteroids, immunomodulatory, antiepileptic and antidepressant drugs usually used in individuals with MS, high incidence of vitamin D deficiency, molecular mechanisms and disuse-loss of mechanical stimuli in bone have an effect on bone integrity (most believe that immobilization of these patients is a minor factor in the etiology of osteoporosis) (Dionyssiotis, 2011).

#### **3.3 Stroke**

Longitudinal studies of body composition in the elderly have shown that body cell mass decreases with age and is lower in women than in men (Steen et al., 1985). A decline in body fat in both the dependent and independent groups nine weeks after admission was found, indicating consumption of energy stores. In contrast, the change of body cell mass between admission and after 9 weeks was significantly greater in the dependent patients compared with the independent (Unosson et al., 1994). Immobilized individuals lose muscle mass irrespective of nutritional intake because of reduced synthesis of proteins, while the rate of breakdown of proteins is unchanged (Schonheyder et al., 1954). During the recovery period the stroke patients seemed to break down body fat to compensate for energy needs, independent of their functional condition. However, change of body cell mass appeared to relate to the patients' functional condition after stroke (Unosson et al., 1994).

A study in 35 stroke patients compared the body composition, including lean tissue mass, fat tissue mass, and bone mineral content, of the paretic leg with that of the non affected leg in patients with stroke and evaluated the effects of time since stroke, spasticity, and motor recovery on the body composition specifically within the first year after stroke found lean tissue mass and bone mineral content of the paretic side to be significantly lower than those of the non affected side; a significant correlation was found between the lean tissue mass and bone mineral content of both the paretic and non affected legs after adjusting for age and weight. On the contrary bone mineral content and lean tissue mass of both the paretic and non affected sides were negatively correlated with time since stroke in patients with stroke for less than 1 year and a higher lean tissue mass and bone mineral content were found in patients with moderate to high spasticity in comparison with patients with low or no spasticity (Celik et al., 2008).

#### **3.4 Cerebral palsy**

Bone mineralization in children with CP has been found lower (bone-mineral values for the total body and total proximal femur) than sex- and age-matched able bodied children. This is illustrated by the BMC Z – scores determined at each skeletal site. The factors that contribute to low bone mineralization include genetic, hormonal, and nutritional problems (especially calcium and vitamin D) and weight-bearing physical activity, oral-motor dysfunction and anticonvulsant medication (Henderson et al., 1995).

Free fat mass (FFM) in cerebral palsy subjects was found significantly lower than that in a normal adolescent population. In 60% of the studied population body fat exceeded the 90th

percentile for age, even if most of the CP children had a low height and weight for age. In female subjects anthropometric measurements were highly correlated with measures of body fatness. Measuring fat by 18O dilution a hydration factor of 0.73 was assumed for FFM. A possible increase in the hydration factor would diminish measured FFM meaning that body fat appears increased. Moreover muscle spasms and spasticity in CP subjects deplete body glycogen. If glycogen is reduced the intracellular water would be reduced and the ratio extracellular water/total body water would increase. The same could result with a loss of body cell mass or an increase in the hydration factor (Bandini et al., 1991).

#### **4. Conclusions**

86 Dual Energy X-Ray Absorptiometry

Endogenous glucocorticoid excess also produces generalized osteoporosis, most prevalent

Beside corticosteroids, immunomodulatory, antiepileptic and antidepressant drugs usually used in individuals with MS, high incidence of vitamin D deficiency, molecular mechanisms and disuse-loss of mechanical stimuli in bone have an effect on bone integrity (most believe that immobilization of these patients is a minor factor in the etiology of osteoporosis)

Longitudinal studies of body composition in the elderly have shown that body cell mass decreases with age and is lower in women than in men (Steen et al., 1985). A decline in body fat in both the dependent and independent groups nine weeks after admission was found, indicating consumption of energy stores. In contrast, the change of body cell mass between admission and after 9 weeks was significantly greater in the dependent patients compared with the independent (Unosson et al., 1994). Immobilized individuals lose muscle mass irrespective of nutritional intake because of reduced synthesis of proteins, while the rate of breakdown of proteins is unchanged (Schonheyder et al., 1954). During the recovery period the stroke patients seemed to break down body fat to compensate for energy needs, independent of their functional condition. However, change of body cell mass appeared to

A study in 35 stroke patients compared the body composition, including lean tissue mass, fat tissue mass, and bone mineral content, of the paretic leg with that of the non affected leg in patients with stroke and evaluated the effects of time since stroke, spasticity, and motor recovery on the body composition specifically within the first year after stroke found lean tissue mass and bone mineral content of the paretic side to be significantly lower than those of the non affected side; a significant correlation was found between the lean tissue mass and bone mineral content of both the paretic and non affected legs after adjusting for age and weight. On the contrary bone mineral content and lean tissue mass of both the paretic and non affected sides were negatively correlated with time since stroke in patients with stroke for less than 1 year and a higher lean tissue mass and bone mineral content were found in patients with moderate to high spasticity in comparison with patients with low or

Bone mineralization in children with CP has been found lower (bone-mineral values for the total body and total proximal femur) than sex- and age-matched able bodied children. This is illustrated by the BMC Z – scores determined at each skeletal site. The factors that contribute to low bone mineralization include genetic, hormonal, and nutritional problems (especially calcium and vitamin D) and weight-bearing physical activity, oral-motor

Free fat mass (FFM) in cerebral palsy subjects was found significantly lower than that in a normal adolescent population. In 60% of the studied population body fat exceeded the 90th

dysfunction and anticonvulsant medication (Henderson et al., 1995).

relate to the patients' functional condition after stroke (Unosson et al., 1994).

in trabecular-rich skeletal regions (Formica et al., 1997).

(Dionyssiotis, 2011).

no spasticity (Celik et al., 2008).

**3.4 Cerebral palsy** 

**3.3 Stroke** 

Other important issues according alterations of body composition are the completeness of lesions (an absence of sensory or motor function below the neurological level, including the lowest sacral segment), because body composition seems to be worst than subjects with incomplete lesions (partial preservation of motor and/or sensory function below the neurological level, including the lowest sacral segment) (Sabo et al., 1991; Demirel et al., 1998; Garland et al., 1992) and aging which contributes to major alterations of body composition.

In disabled subjects the most important issue according to body composition is how to promote optimal body weight to reduce risk of diseases such as coronary heart disease, noninsulin dependent diabetes mellitus, lipid abnormalities and fractures because of bone loss. Dietary changes, individualized physical activity programs and medication should be taken in mind in therapy when we deal with this subgroup of subjects. However, selfmanagement of dietary changes to improve weight control and disease should be the case, which means they need to follow diets with lower energy intake and at the same time to eat regularly foods rich in nutrients (Groah et al., 2009).

We need to take in mind that healthy BMI values often underestimate body fat and may mask the adiposity and spasticity did not defend skeletal muscle mass and bone, supporting the concept that in neurologic disabilities the myopathic muscle could not recognize correctly the stimulation because of the neurogenic injury. Moreover, disabled subjects mostly transfer much of the weight-bearing demands of daily activities to their upper extremities reducing the weight-bearing of the affected paralyzed muscles triggering a cycle of added muscle atrophy which interacts with the continuous catabolic action caused by the neurogenic factor. Finally, an irreversible (once established) decline in bone mineral density, bone mineral content as well as geometric characteristics of bone is expected and the duration of lesion-injury is positively correlated with the degree of bone loss.

Further research about body composition is needed in all physical disabilities and more longitudinal studies to quantitate and monitor body composition changes and to modify our therapeutic interventions. However, prevention rather than treatment may have the greatest potential to alleviate these major complications. Therapies should focus on how to perform weight bearing, standing or therapeutically walking activities early in the rehabilitation program to gain benefits according to muscles and bones.

Body Composition in Disabilities of Central Nervous System 89

Coupaud S, McLean AN, Allan DB. Role of peripheral quantitative computed tomography

Dauty M, Perrouin Verbe B, Maugars Y, Dubois C, Mathe JF. Supralesional and sublesional bone mineral density in spinal cord-injured patients. Bone. 2000;27:305-9. Demirel G, Yilmaz H, Paker N, Onel S. Osteoporosis after spinal cord injury. Spinal Cord.

Dionyssiotis Y, Lyritis GP, Papaioannou N, Papagelopoulos P, Thomaides T. Influence of

Dionyssiotis Y, Trovas G, Galanos A, Raptou P, Papaioannou N, Papagelopoulos P,

Dionyssiotis Y. Changes in bone density and strength of the tibia and alterations of lean and

Dionyssiotis, Y. Bone loss and fractures in multiple sclerosis: focus on epidemiologic and

Dionyssiotis, Y. Spinal cord injury-related bone impairment and fractures: an update on

Dionyssiotis, Y, Lyritis GP, Mavrogenis AF, Papagelopoulos PJ. Factors influencing bone

Elias CF, Lee C, Kelly J, Aschkenasi C, Ahima RS, Couceyro PR, Kuhar MJ, Saper CB,

Ferretti J.L., Cointry G.R., Capozza R.F., Zanchetta J.R. Dual energy X-ray absorptiometry.

T.J.Peters (eds),Greenwich Medical Media, Ltd., London, 2001; p.451-458. Forbes GB, Simon W, Amatruda JM. Is bioimpedance a good predictor of body-composition

Forbes GB. Human body composition: growth, aging, nutrition, and activity. New York:

Formica CA, Cosman F, Nieves J, Herbert J, Lindsay R. Reduced bone mass and fat-free

Frey-Rindova P, de Bruin ED, Stussi E, Dambacher MA, Dietz V. Bone mineral density in

glucocorticoid Use. Calcif Tissue Int. 1997 Aug;61(2):129-33.

of the Musculoskeletal System, University of Athens, Athens, 2008a. Dionyssiotis Υ, Petropoulou Κ, Rapidi CA, Papagelopoulos PJ, Papaioannou N, Galanos A,

http://cirrie.buffalo.edu/encyclopedia/en/article/340/

physiopathological features. Int J Gen Med. 2011b; 4: 505-9.

loss in paraplegia. Hippokratia. 2011a ; 15(1):54-9.

Clinical Densitometry. 2008b;11: 437-43.

Interact. 2011c; 11(3):257-65.

cord. Neuron. 1998;21:1375-85.

change? Am J Clin Nutr 1992;56:4-6.

Springer-Verlag; 1987.

Oct;38(10):989-95.

Dev. 2009;46(8):1037-44.

1998;36:8

in identifying disuse osteoporosis in paraplegia. Skeletal Radiol. 2009

neurological level of injury in bones, muscles, and fat in paraplegia. J Rehabil Res

Petropoulou K, Lyritis GP. Bone loss and mechanical properties of tibia in spinal cord injured men. J Musculoskelet Neuronal Interact. 2007 Jan-Mar;7(1):62-8. Dionyssiotis Y. (2011d). Bone Loss in Spinal Cord Injury and Multiple Sclerosis. In: JH Stone,

M Blouin, editors. International Encyclopedia of Rehabilitation, av. online:

fat mass in chronic paraplegic men. Doctoral Dissertation Laboratory for Research

Papadaki P, and Lyritis GP. Body Composition in Paraplegic Men. Journal of

epidemiology and physiopathological mechanisms. J Musculoskelet Neuronal

Elmquist JK. Leptin activates hypothalamic CART neurons projecting to the spinal

Skeletal Muscle: Pathology, Diagnosis and Management of Disease. V.R.Preedy,

mass in women with multiple sclerosis: effects of ambulatory status and

upper and lower extremities during 12 months after spinal cord injury measured by peripheral quantitative computed tomography. Spinal Cord. 2000;38:26–32.

#### **5. References**


Bandini LG, Schoeller DA, Fukagawa NK, Wykes LJ, Dietz WH. Body composition and

Bauman WA, Spungen AM, Raza M, Rothstein J, Zhang RL, Zhong YG, Tsuruta M, Shahidi

Bauman WA, Spungen AM, Zhong YG, Mobbs CV. Plasma leptin is directly related to body adiposity in subjects with spinal cord injury. Horm Metab Res. 1996;28:732-6. Bauman WA, Spungen AM. 1994 Disorders of carbohydrate and lipid metabolism in

Bender R, Trautner C, Spraul M, Berger M. Assessment of excess mortality in obesity. Am J

Biering-Sorensen F, Bohr HH, Schaadt OP. Longitudinal study of bone mineral content in

Buchholz AC, Bugaresti JM. A review of body mass index and waist circumference as

Castro MJ, Apple DF Jr, Hillegass EA, and Dudley GA. Influence of complete spinal cord

Castro MJ, Apple DF Jr, Staron RS, Campos GE, Dudley GA. Influence of complete spinal

Celik B, Ones K, Ince N. Body composition after stroke. Int J Rehabil Res. 2008 Mar;31(1):93-

Chad KE, McKay HA, Zello GA, Bailey DA, Faulkner RA, Snyder RE. Body composition in

Chantraine A, Nusgens B, Lapiere CM. Bone remodelling during the development of

Chow YW, Inman C, Pollintine P, Sharp CA, Haddaway MJ, el Masry W, Davie

Clarys JP, Martin AD, Drinkwater DT. Gross tissue weights in the human body by cadaver

Clasey JL, Janowiak AL, Gater DR Relationship between regional bone density

Correia ML, Morgan DA, Mitchell JL, Sivitz WI, Mark AL, Haynes WG. Role of

osteoporosis in paraplegia. Calcif Tissue Int. 1986;38:323-7.

energy expenditure in adolescents with cerebral palsy or myelodysplasia.Pediatr

R, Pierson RN Jr, Wang J, et al. Coronary artery disease: metabolic risk factors and latent disease in individuals with paraplegia. Mt Sinai J Med. 1992 Mar;59(2):163-8.

veterans with paraplegia or quadriplegia: A model of premature aging. Metabolism

the lumbar spine, the forearm and the lower extremities after spinal cord injury.

markers of obesity and coronary heart disease risk in persons with chronic spinal

injury on skeletal muscle cross-sectional area within the first 6 months of injury.

cord injury on skeletal muscle within 6 mo of injury. J Appl Physiol. 1999b;86:350-8.

nutritionally adequate ambulatory and non-ambulatory children with cerebral palsy and a healthy reference group. Dev Med Child Neurol. 2000 May;42(5):334-9.

MW.Ultrasound bone densitometry and dual energy X-ray absorptiometry in patients with spinal cord injury: a cross-sectional study. Spinal Cord. 1996

measurements and the time since injury in adults with spinal cord injuries. Arch

corticotrophin-releasing factor in effects of leptin on sympathetic nerve activity and

**5. References** 

Res. 1991 Jan;29(1):70-7.

Epidemiol. 1998;147:42-8.

Eur J Clin Invest. 1990; 20:330-5.

cord injury. Spinal Cord. 2005;43:513-8.

Eur J Appl Physiol 80: 373–378, 1999a.

dissection. Hum Biol. 1984;56:459-73.

arterial pressure. Hypertension. 2001;38:384-8.

Phys Med Rehabil. 2004;85:59–64

.43: 749.756.

6.

Dec;34(12):736-41.


Body Composition in Disabilities of Central Nervous System 91

Kent-Braun JA, Ng AV, Castro M, et al. Strength, skeletal muscle composition and enzyme

Kent-Braun JA, Sharma KR, Weiner MW, Miller RG. Effects of exercise on muscle activation

Krupp, LB, Serafin DJ, Christodoulou C. Multiple sclerosis-associated fatigue. Expert Rev

Kunkel CF, Scremin AM, Eisenberg B, Garcia JF, Roberts S, Martinez S. Effect of ''standing''

Kuperminc MN, Stevenson RD. Growth and nutrition disorders in children with cerebral

LaForgia J, Dollman J, Dale MJ, Withers RT, Hill AM. Validation of DXA body composition estimates in obese men and women. Obesity (Silver Spring). 2009;17:821-6. Lambert CP, Archer RL, Evans WJ. Body composition in ambulatory women with multiple

Laskey MA. Dual-energy X-ray absorptiometry and body composition. Nutrition.1996

Lazo MG, Shirazi P, Sam M, Giobbie-Hurder A, Blacconiere MJ, Muppidi M. Osteoporosis and risk of fracture in men with spinal cord injury. Spinal Cord. 2001;39:208-14. Leslie WD, Nance PW. Dissociated hip and spine demineralization: a specific finding in

Liu LF, Roberts R, Moyer-Mileur L, Samson-Fang L. Determination of body composition in

Lohman, TG.; Roche, AF.; Martorell, R. Anthropometric standardization reference manual.

Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang Y, Fei H, Kim S, Lallone R,

Mahon AK, Flynn MG, Iglay HB, Stewart LK, Johnson CA, McFarlin BK, Campbell WW.

women: comparison of methods. J Nutr Health Aging. 2007;11:203-13. Maimoun L, Fattal C, Micallef JP, Peruchon E, Rabischong P. Bone loss in spinal cordinjured patients: from physiopathology to therapy. Spinal Cord. 2006;44:203-10.

children with cerebral palsy: bioelectrical impedance analysis and anthropometry vs dual-energy x-ray absorptiometry. J Am Diet Assoc. 2005 May;105(5):794-7. Lohman TG. Applicability of body composition techniques and constants for children and

youth. In: Pandolf KB, editor. Exercise and sport sciences reviews. Vol 14. New

Ranganathan S, et al. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat Med. 1995;1:1155-

Measurement of body composition changes with weight loss in postmenopausal

sympathetic preganglionic neurons after spinal cord injury. Spinal Cord. 1999;37:6-

on spasticity, contracture, and osteoporosis in paralyzed males. Arch Phys Med

activity in multiple sclerosis. J Appl Physiol 1997;83:1998-2004.

13.

Neurother. 2010;10:1437-47.

palsy. Dev Disabil Res Rev. 2008;14(2):137-46.

sclerosis. Arch Phys Med Rehabil 2002;83:1559-61.

spinal cord injury. Arch Phys Med Rehabil. 1993; 74:960-4.

Rehabil. 1993;74:73–8.

Jan;12(1):45-51.

61.

York: Macmillan; 1986. p 325-57.

Human Kinetics Books; Champaign: 1988

and metabolism in multiple sclerosis.Muscle Nerve 1994;17:1162-9. Kocina P. Body composition of spinal cord injured adults. Sports Medicine. 1997; 23:48-60. Krassioukov AV, Bunge RP, Pucket WR, Bygrave MA. The changes in human spinal


Frisancho RA. New norms of upper limb fat and muscle areas for assessment of nutritional

Fruhbeck G, Jebb SA, Prentice AM. Leptin: physiology and pathophysiology. Clin Physiol.

Garland DE, Stewart CA, Adkins RH, Hu SS, Rosen C, Liotta FJ, Weinstein DA. 1992.

Gorgey AS, Dudley GA. Skeletal muscle atrophy and increased intramuscular fat after

Groah SL, Nash MS, Ljungberg IH, Libin A, Hamm LF, Ward E, Burns PA, Enfield G.

Hafer-Macko CE, Ryan AS, Ivey FM, Macko RF. Skeletal muscle changes after hemiparetic

Haque MS, Minokoshi Y, Hamai M, Iwai M, Horiuchi M, Shimazu T. Role of the

Henderson RC, Lin PP, Greene WB. (1995). Bone-mineral density in children and

Heymsfield SB, Wang J, Heshka S, Kehayias JJ, Pierson RN. Dual-photon absorptiometry:

Hildreth HG, Johnson RK, Goran MI, Contompasis SH. Body composition in adults with

Horowitz JF, Coppack SW, Paramore D, Cryer PE, Zhao G, Klein S. Effect of short-term

Horowitz JF, Klein S. Whole body and abdominal lipolytic sensitivity to epinephrine is

Ingemann-Hansen T, Halkjaer-Kristensen J. Lean and fat component of the human thigh: the

Jeon JY, Steadward RD, Wheeler GD, Bell G, McCargar L, Harber V. Intact sympathetic

Jiang SD, Dai LY, Jiang LS. Osteoporosis after spinal cord injury. Osteoporos Int.

Jones LM, Goulding A, Gerrard DF. DEXA: a practical and accurate tool to demonstrate total

with spinal cord injury. J Clin Endocrinol Metab. 2003;88:402-7.

established methods. Am J Clin Nutr. 1989 Jun;49(6):1283-9.

technique. Am J Clin Nutr. 1997 Dec;66(6):1436-42.

Nutrient intake and body habitus after spinal cord injury: an analysis by sex and

stroke and potential beneficial effects of exercise intervention strategies. J Rehabil

sympathetic nervous system and insulin in enhancing glucose uptake in peripheral tissues after intrahypothalamic injection of leptin in rats. Diabetes. 1999;48:1706-12.

adolescents who have spastic cerebral palsy.*Journal of Bone and Joint Surgery* 77A:

comparison of bone mineral and soft tissue mass measurements in vivo with

cerebral palsy by dual-energy X-ray absorptiometry, bioelectrical impedance analysis, and skinfold anthropometry compared with the 18O isotope-dilution

fasting on lipid kinetics in lean and obese women. Am J Physiol. 1999;276:E278-

suppressed in upper body obese women. Am J Physiol Endocrinol Metab.

effects of immobilization in plaster and subsequent physical training. *Scand J* 

nervous system is required for leptin effects on resting metabolic rate in people

and regional bone loss, lean tissue loss and fat mass gain in paraplegia. Spinal

Osteoporosis after spinal cord injury. J Orthop Res .10 :371.378.

incomplete spinal cord injury. Spinal Cord 2007;45(4):304–309.

level of injury. J Spinal Cord Med. 2009;32:25-33.

status. Am J Clin Nutr 1981;34:2540–2545.

1998;18:399-419.

Res Dev. 2008;45(2):261-72.

1671–81.

84.

2000;278:E1144-52.

2006;17:180-92.

Cord. 1998;36:637-40

*Rehabil Med.* 1977;9:67–72


Body Composition in Disabilities of Central Nervous System 93

Samson-Fang LJ, Stevenson RD. Identification of malnutrition in children with cerebral

Scelsi R. Skeletal muscle pathology after spinal cord injury. Basic Appl Myol. 2001;11:75-

Schonheyder F, Heilskov NCS, Olesen K. Isotopic studies on the mechanism of negative

Seidell JC, Verschuren WM, van Leer EM, Kromhout D. Overweight, underweight, and

Shields RK, Dudley-Javoroski S. Musculoskeletal adaptations in chronic spinal cord injury:

Shields RK. Muscular, skeletal, and neural adaptations following spinal cord injury. J

Sioka C, Fotopoulos A, Georgiou A, Papakonstantinou S, Pelidou SH, Kyritsis AP, Kalef-

Slaughter MH, Lohman TG, Boileau RA, et al. Skinfold equations for estimation of body

Smeltzer SC, Zimmerman V, Capriotti T. Osteoporosis risk and low bone mineral density in women with physical disabilities. Arch Phys Med Rehabil. 2005;86:582-6. Spungen AM, Adkins RH, Stewart CA, Wang J, Pierson RN Jr, Waters RL, Bauman WA.

Stallings VA, Cronk CE, Zemel BS, Charney EB. Body composition in children with spastic

Steen B, Lundgren BK, Isaksson B. Body composition at age 70, 75, 79, and 81 years: a

Szollar SM, Martin EM, Parthemore JG, Sartoris DJ, Deftos LJ. Densitometric patterns of spinal cord injury associated bone loss. Spinal Cord. 1997 Jun;35(6):374-82. Tsuzuku S, Ikegami Y, Yabe K. Bone mineral density differences between paraplegic and quadriplegic patients: a cross-sectional study. Spinal Cord. 1999; 37:358-61. Uebelhart D, Demiaux-Domenech B, Roth M, Chantraine A. Bone metabolism in spinal cord

Unosson M, Ek AC, Bjurulf P, von Schenck H, Larsson J. Feeding dependence and

Van Der Ploeg GE, Withers RT, Laforgia J. Percent body fat via DEXA: comparison with a

nutritional status after acute stroke. Stroke 1994, 25(2):366-371.

four-compartment model. J Appl Physiol. 2003;94:499-506.

Illness in the Elderly. New York, NY: Pergamon Press, Inc; 1985:49-52. Stevenson RD, Roberts CD, Vogtle L. The effects of non-nutritional factors on growth in

quadriplegic cerebral palsy. J Pediatr. 1995 May;126(5 Pt 1):833-9.

fatness in children and youth. Hum Biol 1988;60:709–723.

sectional study. J Appl Physiol. 2003;95: 2398–2407.

cerebral palsy. Dev Med Child Neurol. 1995;37: 124–130

2000;42:162–168

1996;156:958-63.

Repair. 2007;21:169-79.

Densitom. 2011 Aug 9.

Paraplegia 1995; 33: 669-673.

Orthop Sports Phys Ther. 2002;32:65-74.

85.

188.

palsy: poor performance of weight-for-height centiles. Dev Med Child Neurol.

nitrogen balance produced by immobilization. Scand Clin Lab Invest.1954;6:178-

mortality. A prospective study of 48.287 men and women. Arch Intern Med.

effects of long-term soleus electrical stimulation training. Neurorehabil Neural

Ezra JA. Body composition in ambulatory patients with multiple sclerosis. J Clin

Factors influencing body composition in persons with spinal cord injury: a cross-

longitudinal population study. In: Chandra RK, ed. Nutrition, Immunity and

injured individuals and in others who have prolonged immobilisation. A review.


McDonald CM, Abresch-Meyer AL, Nelson MD, Widman LM. Body mass index and body

Mitsiopoulos N, Baumgartner RN, Heymsfield SB, Lyons W, Gallagher D, and Ross R.

imaging and computerized tomography. J Appl Physiol. 85: 115–122, 1998. Modlesky CM, Bickel CS, Slade JM, Meyer RA, Cureton KJ, Dudley GA. Assessment of

National Institutes of Health Technology Assessment Conference Statement. Bioelectrical

Ng AV, Kent-Braun JA. Quantitation of lower physical activity in persons with multiple

Ogilvie C, Bowker P, Rowley DI. The physiological benefits of paraplegic orthotically aided

Olgiati R, Burgunder JM, Mumenthaler M. Increased energy cost of walking in multiple

Olle MM, Pivarnik JM, Klish WJ, Morrow JR Jr. Body composition of sedentary and

Patrick J, Gisel E. Nutrition for the feeding impaired child. J Neuro Rehab 1990;4:115–119. Peppler WW, Mazess RB. 1981. Total body bone mineral and lean body mass by dual-

Pietrobelli A, Formica C, Wang AM, Heymsfield SB. 1996. Dual-energy X-ray

Rittweger J, Beller G, Ehrig J, Jung C, Koch U, Ramolla J, Schmidt F, Newitt D, Majumdar S,

Ryan AS, Dobrovolny CL, Silver KH, Smith GV, Macko RF. Cardiovascular fitness after

Ryan AS, Dobrovolny CL, Smith GV, Silver KH, Macko RF. Hemiparetic muscle atrophy

Sabo D, Blaich S, Wenz W, Hohmann M, Loew M, Gerner HJ. Osteoporosis in patients

electrical conductivity. Arch Phys Med Rehabil. 1993;74:706-10.

photon absorptiometry. Calcif Tissue Int 33:353-359

Physiol 271 (Endocrinol Metab 34): E941-E951

with spinal cord injury. J Spinal Cord Med. 2007;30:S97-104.

cord injury. Spinal Cord. 2008 Mar;46:192-7.

sclerosis. Med Sci Sports Exerc 1997;29: 517-23.

Institutes of Health, 1994:12-4

1988;69:846-9.

Bone. 2000;27:319-26.

2002;83(12):1703–1707.

2000;9:185–191.

8.

walking. Paraplegia. 1993;31:111-5.

composition measures by dual x-ray absorptiometry in patients aged 10 to 21 years

Cadaver validation of skeletal muscle measurement by magnetic resonance

skeletal muscle mass in men with spinal cord injury using dual-energy X-ray absorptiometry and magnetic resonance imaging. J Appl Physiol. 2004;96:561-5. Mojtahedi MC, Valentine RJ, Arngrímsson SA, Wilund KR, Evans EM. The association

between regional body composition and metabolic outcomes in athletes with spinal

impedance analysis in body composition measurement. Bethesda, MD: National

sclerosis: effect of spasticity, ataxia, and weakness. Arch Phys Med Rehabil

physically active spinal cord injured individuals estimated from total body

absorptiometry body composition model: review of physical concepts. Am J

Schiessl H, Felsenberg D. Bone-muscle strength indices for the human lower leg.

stroke: Role of muscle mass and gait deficit severity. J Stroke Cerebro Dis

and increased intramuscular fat in stroke patients. Arch Phys Med Rehabil

with paralysis after spinal cord injury: a cross sectional study in 46 male patients with dual-energy X-ray absorptiometry. Arch Orthop Trauma Surg. 2001;121:75–


**Part 3** 

**Miscellaneous** 

Wang YH, Huang TS, Liang HW, Su TC, Chen SY, Wang TD. Fasting serum levels of adiponectin, ghrelin, and leptin in men with spinal cord injury. Arch Phys Med Rehabil. 2005;86:1964-8.

## **Part 3**

### **Miscellaneous**

94 Dual Energy X-Ray Absorptiometry

Wang YH, Huang TS, Liang HW, Su TC, Chen SY, Wang TD. Fasting serum levels of

Rehabil. 2005;86:1964-8.

adiponectin, ghrelin, and leptin in men with spinal cord injury. Arch Phys Med

**7** 

*Spain* 

**Internal Design of the Dry Human Ulna by DXA** 

Dual energy X-ray absorptiometry (DXA) has been used to study dry bones such as spine, femur, jaw... to detect the first onsets of the ossification centers; In clinical practice, DXA is widely used for diagnosis and evaluation of osteoporosis, and a new generation of DXA scanners offers software for performing vertebral morphometry analysis (Blake & Fogelman, 1997). Also, many bone analyses have been performed on experimental animals using DXA

Studies on the spatial distribution of bone mineral density (BMD) in the whole bone, reflecting its morphological pattern are scarce (Gómez-Pellico et al., 1993 & Fernández-Camacho et al., 1996). In addition, there are just few studies regarding the anthropometric

In order to improve the treatment of the elbow's injury, knowledge related to the resistance of the bone is important to understand the origin of the fractures as well as to improve elbow fracture recovery (Heep, 2007). Most studies investigate the humeral component,

In order to develop an implant that carries out the mechanical characteristics of a native bone, we must study the trabecular architecture of the human ulna proximal extremity.

The ulna is a long bone, placed at the medial side of the forearm, parallel to the radius. It is divisible into a body and two extremities. Its upper extremity, of great thickness and strength, forms a large part of the elbow-joint; the bone diminishes in size from above downward, its lower extremity being very small, and excluded from the wrist-joint by the

The upper extremity presents two curved processes, the olecranon and the coronoid process; and two concave, articular cavities, the trochlear and radial notches. The olecranon is a large, thick, curved eminence, situated at the upper and back part of the ulna. The coronoid process is a triangular eminence projecting forward from the upper and front part of the ulna. Its base is continuous with the body of the bone, and of

interposition of an articular disk (the ulna articulates with the humerus and radius).

**1. Introduction** 

(Tsujio et al., 2009).

characteristics of the human ulna (Weber et al., 2009).

**1.1 Brief anatomy of the human ulna** 

considerable strength.

while the ulna component is not being studied as much (Goto, 2009).

S. Aguado-Henche, A. Bosch-Martín,

*University of Alcalá* 

P. Spottorno-Rubio and R. Rodríguez-Torres

### **Internal Design of the Dry Human Ulna by DXA**

S. Aguado-Henche, A. Bosch-Martín, P. Spottorno-Rubio and R. Rodríguez-Torres *University of Alcalá Spain* 

#### **1. Introduction**

Dual energy X-ray absorptiometry (DXA) has been used to study dry bones such as spine, femur, jaw... to detect the first onsets of the ossification centers; In clinical practice, DXA is widely used for diagnosis and evaluation of osteoporosis, and a new generation of DXA scanners offers software for performing vertebral morphometry analysis (Blake & Fogelman, 1997). Also, many bone analyses have been performed on experimental animals using DXA (Tsujio et al., 2009).

Studies on the spatial distribution of bone mineral density (BMD) in the whole bone, reflecting its morphological pattern are scarce (Gómez-Pellico et al., 1993 & Fernández-Camacho et al., 1996). In addition, there are just few studies regarding the anthropometric characteristics of the human ulna (Weber et al., 2009).

In order to improve the treatment of the elbow's injury, knowledge related to the resistance of the bone is important to understand the origin of the fractures as well as to improve elbow fracture recovery (Heep, 2007). Most studies investigate the humeral component, while the ulna component is not being studied as much (Goto, 2009).

In order to develop an implant that carries out the mechanical characteristics of a native bone, we must study the trabecular architecture of the human ulna proximal extremity.

#### **1.1 Brief anatomy of the human ulna**

The ulna is a long bone, placed at the medial side of the forearm, parallel to the radius. It is divisible into a body and two extremities. Its upper extremity, of great thickness and strength, forms a large part of the elbow-joint; the bone diminishes in size from above downward, its lower extremity being very small, and excluded from the wrist-joint by the interposition of an articular disk (the ulna articulates with the humerus and radius).

The upper extremity presents two curved processes, the olecranon and the coronoid process; and two concave, articular cavities, the trochlear and radial notches. The olecranon is a large, thick, curved eminence, situated at the upper and back part of the ulna. The coronoid process is a triangular eminence projecting forward from the upper and front part of the ulna. Its base is continuous with the body of the bone, and of considerable strength.

Internal Design of the Dry Human Ulna by DXA 99

To begin the scan, (figure 1) the starting point was placed 0, 5 cm directly above the upper extremity. A baseline point was marked under the lower extremity. A third point (goal line)

This technique has high accuracy and precision, approaching 1%. The speed of scanning was 60 mms, with an interlinear space of 1 mm and point by point resolution of 1 mm horizontal x 1 mm vertical. The defined exploration was completed as outlined in an average time of 10-15

min. Scan acquisition and scan analyses were performed by one investigator (figure 2).

was marked 1 cm from the more lateral part of the bone.

Fig. 1. Definition of the exploration area

Fig. 2. Densitometric image of the ulna.

Dry ulna calculations were performed for the following magnitudes:

**BMD**: Bone mineral density, in grams / cm2.

**BMC**: Bone mineral content, in grams. BMC is defined as the mass of

dependent parameter (Schoenau, 2004).

mineral contained in an entire bone or as the mass of mineral per unit bone length. Bone mineral content is obviously a size-

Its antero-inferior surface is concave, and marked by a rough impression for the insertion of the brachialis muscle. The trochlear notch is a large depression, formed by the olecranon and the coronoid process, and serving for articulation with the trochlea of the humerus. The notch is concave from above downward, and divided into a medial and a lateral portion by a smooth ridge running from the summit of the olecranon to the tip of the coronoid process. The radial notch is a narrow, oblong, articular depression on the lateral side of the coronoid process; it receives the circumferential articular surface of the head of the radius. The lower extremity of the ulna is small, and presents two eminences; the lateral and larger is a rounded, articular eminence, termed the head of the ulna; the medial, narrower and more projecting is a non-articular eminence named the styloid process.

The ulna is ossified from three centers: one for the body, the inferior extremity, and the top of the olecranon. Ossification begins about the eighth week of fetal life. About the fourth year, a center appears in the middle of the ulnar head, and soon extends into the styloid process. About the tenth year, a center appears in the olecranon near its extremity. The upper epiphysis joins the body about the sixteenth year and the lower about the twentieth.

#### **2. Objective**

In this chapter, we set out to show, by means of densitometric analysis with dual energy Xray absorptiometry (DXA) the internal design of the human ulna, to verify that the bone tissue distribution is not homogeneous and that this corresponds to the trabecular architecture of the bone.

#### **3. Material and method**

A random sample of 41 dry right ulnas from the skeletal collection of the Anatomy and Embriology Department of the University of Alcala was studied excluding those bones which presented any alterations or damage. A Norland XR-26 densitometer, software 2.5 (Norland Co., Fort Atkinson, WI, USA; Emsor SA, Madrid) was used for all studies. Each scan session was preceded by a calibration routine using a standard calibration block supplied by the manufacturer.

The bone is placed well centred on the examining board. It is important to check for stability so as not to vary its position during the study. Cotton gauze may be needed for an optimal stabilization. The bones are exposed directly, without any water or other materials that may resemble soft tissue.

For the densitometric analysis of the human ulna structure two projections were performed: lateral and antero-posterior.

For the study in two positions, the reference will be the ridge of the trochlear notch of the epiphyseal ulna (incisura trochlearis) which corresponds to the throat of the trochlea – humerus- (Gómez-Oliveros, 1962).


Its antero-inferior surface is concave, and marked by a rough impression for the insertion of the brachialis muscle. The trochlear notch is a large depression, formed by the olecranon and the coronoid process, and serving for articulation with the trochlea of the humerus. The notch is concave from above downward, and divided into a medial and a lateral portion by a smooth ridge running from the summit of the olecranon to the tip of the coronoid process. The radial notch is a narrow, oblong, articular depression on the lateral side of the coronoid process; it receives the circumferential articular surface of the head of the radius. The lower extremity of the ulna is small, and presents two eminences; the lateral and larger is a rounded, articular eminence, termed the head of the ulna; the medial, narrower and more

The ulna is ossified from three centers: one for the body, the inferior extremity, and the top of the olecranon. Ossification begins about the eighth week of fetal life. About the fourth year, a center appears in the middle of the ulnar head, and soon extends into the styloid process. About the tenth year, a center appears in the olecranon near its extremity. The upper epiphysis joins the body about the sixteenth year and the lower about the twentieth.

In this chapter, we set out to show, by means of densitometric analysis with dual energy Xray absorptiometry (DXA) the internal design of the human ulna, to verify that the bone tissue distribution is not homogeneous and that this corresponds to the trabecular

A random sample of 41 dry right ulnas from the skeletal collection of the Anatomy and Embriology Department of the University of Alcala was studied excluding those bones which presented any alterations or damage. A Norland XR-26 densitometer, software 2.5 (Norland Co., Fort Atkinson, WI, USA; Emsor SA, Madrid) was used for all studies. Each scan session was preceded by a calibration routine using a standard calibration block

The bone is placed well centred on the examining board. It is important to check for stability so as not to vary its position during the study. Cotton gauze may be needed for an optimal stabilization. The bones are exposed directly, without any water or other materials that may

For the densitometric analysis of the human ulna structure two projections were performed:

For the study in two positions, the reference will be the ridge of the trochlear notch of the epiphyseal ulna (incisura trochlearis) which corresponds to the throat of the trochlea –

Anteroposterior Position: The axis of the ridge of the trochlear notch is perpendicular to

Lateral Position: The axis of the ridge of the trochlear notch is parallel to the axis of the

projecting is a non-articular eminence named the styloid process.

**2. Objective** 

architecture of the bone.

**3. Material and method** 

supplied by the manufacturer.

lateral and antero-posterior.

examining board.

humerus- (Gómez-Oliveros, 1962).

the axis of the examining board.

resemble soft tissue.

To begin the scan, (figure 1) the starting point was placed 0, 5 cm directly above the upper extremity. A baseline point was marked under the lower extremity. A third point (goal line) was marked 1 cm from the more lateral part of the bone.

Fig. 1. Definition of the exploration area

This technique has high accuracy and precision, approaching 1%. The speed of scanning was 60 mms, with an interlinear space of 1 mm and point by point resolution of 1 mm horizontal x 1 mm vertical. The defined exploration was completed as outlined in an average time of 10-15 min. Scan acquisition and scan analyses were performed by one investigator (figure 2).

Fig. 2. Densitometric image of the ulna.

Dry ulna calculations were performed for the following magnitudes:


Internal Design of the Dry Human Ulna by DXA 101

Projection LAT Mínimum Máximum Mean SD Total BMD 0,38 0,94 0,67 0,14 Total BMC 13,15 46,74 28,72 8,50 Total Area 32,81 52,51 41,84 5,25 Total Lenght 20,40 28,05 24,50 1,66 Total Width 3,15 6,60 4,08 0,59

Table 2. Descriptive statistics of the total ulna in projection lateral (n=41). SD: Standard

AP-Area 10,58 1,57 9,08 1,33 8,12 1,42 7,02 1,01 6,22 0,94 AP-Lenght 4,88 0,33 4,88 0,33 5,03 0,41 4,88 0,33 4,88 0,33 LAT-Area 11,68 1,52 8,94 1,24 8,09 1,07 6,98 0,82 6,24 1,00 LAT-Lenght 4,88 0,33 4,88 0,33 5,03 0,43 4,88 0,33 4,88 0,33 Table 3. Descriptive statistics of the regions of interest in projections antero-posterior (AP)

Table 4. BMD (in grams/cm2) and BMC (in grams) of the regions of interest.

Dual energy X-ray absorptiometry (DXA) allows us to gather quantitative information on bone mineral content (BMC) and bone mineral density (BMD) of the bone (Wahner et al., 1985). As previously reported (Hvid et al., 1985), there is a close relationship between bone

In literature, there are various studies on long bones, such as the femur, the tibia, the humerus and the radius (Wahner et al., 1985; Kawashima & Uhthoff, 1991; Gómez-Pellico et

ROI 1 ROI 2 ROI 3 ROI 4 ROI 5 Media SD Media SD Media SD Media SD Media SD

deviation.

**5. Discussion** 

mass and bone strength.

and lateral (LAT). n=41. SD: Standard deviation.


For the purpose of this survey, in both projections five equal regions of interest (ROI) were selected: proximal (ROI-1), proximal-intermediate (ROI-2), intermediate (ROI-3), distalintermediate (ROI-4) and distal (ROI-5). The total region corresponded to the area of the full length and height of the bone (figure 3).

All statistical calculations were performed using Statgraphics Plus (version 5.1) and SPSS (Statistical Package for Social Sciences), version 15.0. The means and standard deviation (SD) for bone mineral density (BMD) and bone mineral content (BMC) were calculated. The bone densities and the bone contents of the various regions of the ulna in the 2 projections were compared by Student's *t* test for paired samples.

Fig. 3. Regions of interest.

#### **4. Results**

DXA indicates that the higher BMD is in the proximal-intermediate region (R2), which is the part of the ulna that bears the higher force of traction. The higher BMC is found in the proximal region (R1) which corresponds to the coronoid process. Lower BMD and BMC are found in the distal region (R5). The total BMD shows significant statistical differences (p ≤ 0.001), which indicates the heterogeneous nature of the distribution of bone mass in the studied bone.

In tables 1 - 4 we present the statistic descriptions of the densitometry variables studied in both.


Table 1. Descriptive statistics of the total ulna in projection antero-posterior (n=41). SD: Standard deviation.

For the purpose of this survey, in both projections five equal regions of interest (ROI) were selected: proximal (ROI-1), proximal-intermediate (ROI-2), intermediate (ROI-3), distalintermediate (ROI-4) and distal (ROI-5). The total region corresponded to the area of the full

All statistical calculations were performed using Statgraphics Plus (version 5.1) and SPSS (Statistical Package for Social Sciences), version 15.0. The means and standard deviation (SD) for bone mineral density (BMD) and bone mineral content (BMC) were calculated. The bone densities and the bone contents of the various regions of the ulna in the 2 projections

DXA indicates that the higher BMD is in the proximal-intermediate region (R2), which is the part of the ulna that bears the higher force of traction. The higher BMC is found in the proximal region (R1) which corresponds to the coronoid process. Lower BMD and BMC are found in the distal region (R5). The total BMD shows significant statistical differences (p ≤ 0.001), which indicates the heterogeneous nature of the distribution of bone mass in the

In tables 1 - 4 we present the statistic descriptions of the densitometry variables studied in

Projection A-P Mínimum Máximum Mean SD Total BMD 0,40 0,97 0,69 0,14 Total BMC 13,10 46,40 28,75 8,63 Total Area 30,73 51,54 41,05 5,70 Total Lenght 20,40 27,90 24,51 1,66 Total Width 2,55 4,35 3,35 0,40

Table 1. Descriptive statistics of the total ulna in projection antero-posterior (n=41). SD:

**AREA**: Measured area, in square centimetres (cm2). **LENGTH**: Total length of the bone, in centimetres (cm). **WIDTH**: Total width of the bone, in centimetres (cm).

length and height of the bone (figure 3).

Fig. 3. Regions of interest.

**4. Results** 

studied bone.

Standard deviation.

both.

were compared by Student's *t* test for paired samples.


Table 2. Descriptive statistics of the total ulna in projection lateral (n=41). SD: Standard deviation.


Table 3. Descriptive statistics of the regions of interest in projections antero-posterior (AP) and lateral (LAT). n=41. SD: Standard deviation.


Table 4. BMD (in grams/cm2) and BMC (in grams) of the regions of interest.

#### **5. Discussion**

Dual energy X-ray absorptiometry (DXA) allows us to gather quantitative information on bone mineral content (BMC) and bone mineral density (BMD) of the bone (Wahner et al., 1985). As previously reported (Hvid et al., 1985), there is a close relationship between bone mass and bone strength.

In literature, there are various studies on long bones, such as the femur, the tibia, the humerus and the radius (Wahner et al., 1985; Kawashima & Uhthoff, 1991; Gómez-Pellico et

Internal Design of the Dry Human Ulna by DXA 103

The human ulna presents a heterogeneous distribution of the BMD. This study confirms that the higher mechanical requirements in the ulna are in the proximal extremity. The differences found in the ulna BMD allows us a better understanding of the construction systematics and their functional activity. We conclude that bone densitometry, measured by the DXA technique, is useful for assessing trabecular architecture of the human skeleton. This study may provide some useful information on plate application for the treatment of

Blake, GM. & Fogelman I. (1997). Technical principles, *Seminars in Nuclear Medicine*

D'Amelio, P., Panattoni, GL. & Isaia GC. (2002). Densitometric study of human developing

Chen, WC., Hsu, WY & Wu, JJ (1991). Stress fracture of the diaphysis of the ulna,

Fernández Camacho, FJ., Morante Martínez, P., Rodríguez Torres, R., Cortés García, A. &

Fowler, K. & Chung, C. (2006). Normal MR imaging anatomy of the elbow, *Radiologic Clinics* 

Gómez Pellico, L., Morante Martínez, P., & Dankloff Mora, C. (1993). Definición

Gómez-Oliveros, L. (1962). *Lecciones de Anatomía Humana. Osteología. Tercera parte. Miembros.*

Goto, A., Murase, T., Hashimoto, J., Oka, K., Yoshikawa, H. & Sugamoto, K. (2009).

Griffith, JF., Genant, HK. (2008). Bone mass and architecture determination: state of the art, *Best practice & Research Clinical Endocrinology & Metabolism.*22(5):737-764. Hepp, P., Josten, C. (2007). Biology and Biomechanics in Osteosynthesis of Proximal

Hvid, I., Jensen, NC., Bünger, C., Solund, K., Djurhuus, JC. (1985). Bone mineral assay: its

Kalkwarf, HJ., Laor, T & Bean JA. (2011). Fracture risk in children with forearm injury is

Kawashima, T. & Uhthoff HK. (1991). Pattern of bone loss of the proximal femur: a

Kim, JM., Mudgal, CS., Konopka, JF & Júpiter JB. (2011). Complications of total elbow arthroplasty, *Journal of American Academy of Orthopaedic Surgeons.* 19(6):328-339.

areal bone density (by DXA), *Osteoporosis International* 22:607-616.

Gómez Pellico, L. (1996). Densitometric analysis of the human calcaneus, *Journal of* 

densitométrica de la morfología estructural del huesos del esqueleto humano, *Jano*

Morphologic análisis of the medulary canal in rheumatoid elbows, *Journal of* 

Humerus Fractures, *European Journal of Trauma and Emergency Surgery* 33(4):337-344.

relation to the mechanical strength of cancellous bone, *Engineering in Medicine*

associated with volumetric bone density and cortical area (by peripheral QCT) and

radiologic, densitometric, and histomorphometric study, *Journal of Orthopaedic* 

dry bones: a review, *Journal of Clinical Densitometric* 5(1):73-78.

*International Orthopaedics* 15: 197-198.

*Shoulder and Elbow Surgery.*18(1):33-37.

**6. Conclusions** 

the elbow injuries.

**7. References** 

27(3):210-228.

*Anatomy* 189:205-209.

XLV:637-640.

14:79-83.

*Research* 9(5):630-640.

Madrid. Marban.

*of North America* 44(4):553-567.

al., 1993; D'Amelio et al., 2002) but we haven't found any references that study the distribution of the BMD in the ulna that describes it´s construction systematics .

According to Wolf's law (Viladot , 2001), the bone adapts its size, shape and structure to the mechanical requirements it receives. Furthermore, Pauwels (Pauwels, 1980; Miralles, 1998) suggests that the mass of the cortical bone is distributed along its axis proportionally to the amount of tensions it receives. In our analysis, we find a wider BMD in the intermediate-proximal region (ROI-2), which corresponds to the region of the bone exposed to the mechanical flexions and to the transmission of weight charges while the superior member is in the extended position. Three different soft tissue structures insert in or attach to the coronoid process, the articular capsule, the tendon of the brachialis muscle, and the anterior band of the ulnar collateral ligament (Fowler & Chung, 2006). Furthermore, the transmission of weight charges travels through the coronoid apophysis, situated in the proximal region, which present the wider BMC in both projections, however, DXA is unable to distinguish between cortical and trabecular bone (Griffith & Genant, 2008).

This and other similar studies will contribute to a better understanding of stress related fractures which are quite scarce in the ulna and cannot easily be found in literature (Chen, WC et al., 1991). Most fractures occur in the middle third of the diaphysis and surrounding areas as a result of mechanical stressing forces of the forearm in a specific position, especially in athletes (Rettig, 1983).

Some authors have agreed on the homogeneous nature of the different diaphysarial regions of the long bones that they study (femur, humerus and tibia) (Gómez Pellico et al., 1993; Fernández-Camacho et al., 1996). As far as the ulna is concerned, BMD displays a more heterogeneous distribution, since we find statistical differences in all studied regions and on the entire bone.

In studies of the dry femur with DXA, epiphysiary regions are those with less BMD which, added to the mechanical requirements of the physiology of the articulation, would explain how hip ostheoporotic fractures occur (Gómez-Pellico et al., 1993). Furthermore, most studies with DXA are based on BMD variations related to loss of bone mass of pathologic nature (McCarthy et al., 1991). This also happens with "in vivo" studies of the radius. Due to the high rate fractures of the distal radius in children, the use this bone to measure BMD, is increasing, essentially as thus to predict the risk of fracture (Kalkwarf et al., 2011). The study of the dry radius with DXA would define its construction systematics.

The results obtained with the DXA technique showed that BMD agrees with the arrangement of the trabecular system in the human ulna, previously described by some authors (Testut & Latarjet, 1949; Gómez-Oliveros, 1962).

In addition, fractures of the coronoid process are rarely seen as an isolated injury. They are encountered more frecuently in association with radial head fractures (Weber et al., 2009).

Due to frequent complications associated with reconstructive surgery for the elbow, implant loosening, periprosthetic fracture, implant failure… (Kim, 2011), that remains higher than arthroplasty of other joints (Sanchez-Sotelo, 2011), the findings that result from this study could contribute to the improvement of elbow prosthesis.

#### **6. Conclusions**

102 Dual Energy X-Ray Absorptiometry

al., 1993; D'Amelio et al., 2002) but we haven't found any references that study the

According to Wolf's law (Viladot , 2001), the bone adapts its size, shape and structure to the mechanical requirements it receives. Furthermore, Pauwels (Pauwels, 1980; Miralles, 1998) suggests that the mass of the cortical bone is distributed along its axis proportionally to the amount of tensions it receives. In our analysis, we find a wider BMD in the intermediate-proximal region (ROI-2), which corresponds to the region of the bone exposed to the mechanical flexions and to the transmission of weight charges while the superior member is in the extended position. Three different soft tissue structures insert in or attach to the coronoid process, the articular capsule, the tendon of the brachialis muscle, and the anterior band of the ulnar collateral ligament (Fowler & Chung, 2006). Furthermore, the transmission of weight charges travels through the coronoid apophysis, situated in the proximal region, which present the wider BMC in both projections, however, DXA is unable to distinguish between cortical and trabecular bone (Griffith &

This and other similar studies will contribute to a better understanding of stress related fractures which are quite scarce in the ulna and cannot easily be found in literature (Chen, WC et al., 1991). Most fractures occur in the middle third of the diaphysis and surrounding areas as a result of mechanical stressing forces of the forearm in a specific position,

Some authors have agreed on the homogeneous nature of the different diaphysarial regions of the long bones that they study (femur, humerus and tibia) (Gómez Pellico et al., 1993; Fernández-Camacho et al., 1996). As far as the ulna is concerned, BMD displays a more heterogeneous distribution, since we find statistical differences in all studied regions and on

In studies of the dry femur with DXA, epiphysiary regions are those with less BMD which, added to the mechanical requirements of the physiology of the articulation, would explain how hip ostheoporotic fractures occur (Gómez-Pellico et al., 1993). Furthermore, most studies with DXA are based on BMD variations related to loss of bone mass of pathologic nature (McCarthy et al., 1991). This also happens with "in vivo" studies of the radius. Due to the high rate fractures of the distal radius in children, the use this bone to measure BMD, is increasing, essentially as thus to predict the risk of fracture (Kalkwarf et al., 2011). The study

The results obtained with the DXA technique showed that BMD agrees with the arrangement of the trabecular system in the human ulna, previously described by some

In addition, fractures of the coronoid process are rarely seen as an isolated injury. They are encountered more frecuently in association with radial head fractures (Weber et al., 2009).

Due to frequent complications associated with reconstructive surgery for the elbow, implant loosening, periprosthetic fracture, implant failure… (Kim, 2011), that remains higher than arthroplasty of other joints (Sanchez-Sotelo, 2011), the findings that result from this study

of the dry radius with DXA would define its construction systematics.

authors (Testut & Latarjet, 1949; Gómez-Oliveros, 1962).

could contribute to the improvement of elbow prosthesis.

distribution of the BMD in the ulna that describes it´s construction systematics .

Genant, 2008).

the entire bone.

especially in athletes (Rettig, 1983).

The human ulna presents a heterogeneous distribution of the BMD. This study confirms that the higher mechanical requirements in the ulna are in the proximal extremity. The differences found in the ulna BMD allows us a better understanding of the construction systematics and their functional activity. We conclude that bone densitometry, measured by the DXA technique, is useful for assessing trabecular architecture of the human skeleton. This study may provide some useful information on plate application for the treatment of the elbow injuries.

#### **7. References**


**8** 

*Japan* 

Kazuhiro Imai1,2,3

*Ex Vivo* **and** *In Vivo* **Assessment of Vertebral** 

**by Dual Energy X-Ray Absorptiometry** 

*2Department of Orthopaedic Surgery, School of Medicine, Tokyo University,* 

*3Department of Orthopaedic Surgery, Tokyo Metropolitan Geriatric Medical Center,* 

Osteoporosis is defined as a skeletal disorder characterized by loss of bone mass, decreased bone strength and resulting in increased risk of bone fracture. The disease is progressive with age, especially in postmenopausal women [1]. Osteoporotic hip fractures and vertebral fractures have become a major social problem because the elderly population continues to increase. Hip fractures account for about 10% of all osteoporosis-related fractures [2]. Hip fractures are particularly devastating and have a particularly negative impact on morbidity. Survivors often suffer severe and prolonged physical and social limitations, and fail to recover normal activity [3]. Vertebral fractures affect approximately 25% of postmenopausal women [4]. Vertebral fractures can be associated with chronic disabling pain and incur loss

In addition to this increased awareness of osteoporosis as a significant health problem, there has been the emergence of several novel drugs that appear to be effective at reducing the risk of fracture, such as bisphosphonates. Consequently, clinicians and researchers are emphasizing the importance of early detection of osteoporosis, aggressive fracture prevention, and monitoring of patients who have high risk of fractures. Fracture risk associated with osteoporosis consisted of bone strength reduction and tendency to fall, therefore it is essential to measure bone strength to assess the risk of fracture. Bone strength reflects the integration of bone density and bone quality, which are influenced by bone

Traditionally, measurement of areal bone mineral density (aBMD) by dual energy X-ray absorptiometry (DXA) has served as the means by which to best diagnose osteoporosis and evaluate fracture risk [6]. In 1994, the World Health Organization (WHO) published a set of diagnostic criteria to define osteoporosis in postmenopausal Caucasian women, using aBMD values measured by DXA [7]. Measurement of aBMD by DXA has been the standard method for diagnosing osteoporosis, in addition to assessing fracture risk and therapeutic effects. However, a variety of problems exist with DXA, which include its relatively high cost, the absence of DXA in many communities, especially in less-developed countries.

architecture, bone turnover, accumulation of damage, and mineralization [5].

**1. Introduction** 

of normal activity.

*1Department of Orthopaedic Surgery, Mishuku Hospital, Tokyo,* 

**Strength and Vertebral Fracture Risk Assessed** 


### *Ex Vivo* **and** *In Vivo* **Assessment of Vertebral Strength and Vertebral Fracture Risk Assessed by Dual Energy X-Ray Absorptiometry**

Kazuhiro Imai1,2,3

*1Department of Orthopaedic Surgery, Mishuku Hospital, Tokyo, 2Department of Orthopaedic Surgery, School of Medicine, Tokyo University, 3Department of Orthopaedic Surgery, Tokyo Metropolitan Geriatric Medical Center, Japan* 

#### **1. Introduction**

104 Dual Energy X-Ray Absorptiometry

McCarthy, CK., Steinberg, GG., Agren, M., Leahey, D., Wyman, E., & Baran DT (1991).

Rettig, AC. (1983). Stress fracture of the ulna in an adolescent tournament tennis player.

Sanchez-Sotelo, J. (2011). Total elbow arthroplasty, *The Open Orthopaedics Journal* 16(5):115-

Schoenau, E., Land, C., Stabrey, A., Remer, T. & Kroke, A. (2004). The bone mass concept: problems in short stature, *European Journal of Endocrinology* 151:S87-S91. Testut, L. & Latarjet, A. (1949). *Tratado de Anatomía Humana. Volume 1* Barcelona: Salvat. Tsujio, M., Mizorogi, T., Kitamura, I., Maeda, Y., Nishijima, K., Kuwahara, S., Ohno, T.,

Viladot Voegeli, A. (2001) *Lecciones Básicas de Biomecánica del Aparato Locomotor.* Barcelona:

Wahner, HW., Eastell, R. & Riggs, BL. (1985). Bone mineral density of the radius: Where do

Weber, MF., Barbosa, DM., Belentani, C., Ramos, PM., Trudell, D. & Resnick, D. (2009).

we stand?, *The Journal of Nuclear Medicine* 26(11):1339-1341.

Niida, S., Nagoya, M., Saito, R. & Tanaka, S. (2009). Bone mineral analisis through dual energy X-ray absorptiometry in laboratory animals, *Journal of Veterinary* 

Coronoid process of the ulna: paleopathologic and anatomic study with imaging correlation. Emphasis on the anteromedial "facet", *Skeletal Radiology* 38(1):61-67.

Miralles Marrero, R. (1998). *Biomecánica Clínica del Aparato Locomotor.* Barcelona: Masson. Pauwels, F. (1980) *Biomechanics of the Locomotor Apparatus. Contribution on the functional of the* 

*of Bone and Joint Surgery* 73(5):774-778.

*Medical Science* 71(11):1493-1497.

123.

Springer.

*Locomotor Apparatus.* Nueva York: Srpinger.

*American Journal of Sports Medicine* 11:103-106.

Quantifying bone loss from the proximal femur after total hip arthroplasty, *Journal* 

Osteoporosis is defined as a skeletal disorder characterized by loss of bone mass, decreased bone strength and resulting in increased risk of bone fracture. The disease is progressive with age, especially in postmenopausal women [1]. Osteoporotic hip fractures and vertebral fractures have become a major social problem because the elderly population continues to increase. Hip fractures account for about 10% of all osteoporosis-related fractures [2]. Hip fractures are particularly devastating and have a particularly negative impact on morbidity. Survivors often suffer severe and prolonged physical and social limitations, and fail to recover normal activity [3]. Vertebral fractures affect approximately 25% of postmenopausal women [4]. Vertebral fractures can be associated with chronic disabling pain and incur loss of normal activity.

In addition to this increased awareness of osteoporosis as a significant health problem, there has been the emergence of several novel drugs that appear to be effective at reducing the risk of fracture, such as bisphosphonates. Consequently, clinicians and researchers are emphasizing the importance of early detection of osteoporosis, aggressive fracture prevention, and monitoring of patients who have high risk of fractures. Fracture risk associated with osteoporosis consisted of bone strength reduction and tendency to fall, therefore it is essential to measure bone strength to assess the risk of fracture. Bone strength reflects the integration of bone density and bone quality, which are influenced by bone architecture, bone turnover, accumulation of damage, and mineralization [5].

Traditionally, measurement of areal bone mineral density (aBMD) by dual energy X-ray absorptiometry (DXA) has served as the means by which to best diagnose osteoporosis and evaluate fracture risk [6]. In 1994, the World Health Organization (WHO) published a set of diagnostic criteria to define osteoporosis in postmenopausal Caucasian women, using aBMD values measured by DXA [7]. Measurement of aBMD by DXA has been the standard method for diagnosing osteoporosis, in addition to assessing fracture risk and therapeutic effects. However, a variety of problems exist with DXA, which include its relatively high cost, the absence of DXA in many communities, especially in less-developed countries.

*Ex Vivo* and *In Vivo* Assessment of Vertebral Strength

introduced by fat within the vertebral bone marrow.

widespread use in studies of osteoporosis.

have sustained a prior fracture [7].

part-body examinations being only a few microsieverts (µSv).

and Vertebral Fracture Risk Assessed by Dual Energy X-Ray Absorptiometry 107

the appendicular skeleton. The development of dual-photon absorptiometry (DPA) and, more recently, dual-energy X-ray absorptiometry (DXA) have resolved at least some of these problems. The different thickness of soft tissue can be accommodated by simultaneous measurement of the transmission of gamma-rays of two different energies, which makes the techniques applicable to any part of the body, but particularly the lumbar spine and hip.

The theory underlying DPA and DXA requires that there are only two components present – bone and soft tissue of uniform composition. In practice, fat forms a further component with attenuation characteristics that differ from those of water, muscle and most organs. A uniform layer of fat is unimportant, but fat is distributed non-uniformly in the region of the lumbar spine and may cause errors of up to 10% in spinal bone mineral. Errors can also be

Total body bone mineral can be measured by DPA, but instrumental problems are greater because of the wide range of count rates and the non-uniform distribution of fat, which introduces errors. However, total body bone mineral measured by neutron activation analysis. As with SPA, the radiation dose for DPA is low, the effective dose equivalent for

Recently, sources of gamma radiation have been replaced by X-ray generators. The necessary pairs of effective energies can be obtained either by K-edge filtering, using cerium or samarium, or by rapidly switching the generator potential. The advantages of these approaches are a higher beam intensity and therefore faster scan, improved spatial resolution with easier identification of vertebrae, and better precision. The absence of source decay also eliminates problems associated with decreasing count rates over the lifetime of the source.

Like DPA, DXA determines bone mineral density from an anterior-posterior image. The sites most commonly measured are the lumbar spine, generally L2-L4, including the intervertebral discs. Other sites include the hip, forearm, whole body and skeletal segments. The error in reproducibility *in vitro* is 1-2%. DXA has been reported to have a high shortterm and long-term precision *in vivo*, which is about twice that of DPA. This has led to its

A recent development has been scanning of the lumbar spine in the lateral position, which has the advantage of eliminating the posterior arch and the spines of the vertebrae as well as aortic calcification from the measurement. Its limitations are the increased soft tissue mass and overlap of the projected image by the ribs and pelvis, so that only one or two vertebrae are measured. Lateral scanning provides a measurement of vertebral depth which, together with the antero-posterior area, can provide a volumetric measurement for calculating bone mineral mass per unit volume. Whether this volumetric density measure is a better predictor of fracture is unknown. The technique may be useful in assessment of bone density in children, allowing accurate assessment of vertebral size. The precision error of measurement of the vertebral body and mid-slice *in vivo* is of the order of 2% [15]. DXA has now largely replaced DPA for screening because of its greater precision, ease of use and freedom from several technical artifacts. The WHO defines osteoporosis as a value for aBMD by DXA 2.5 standard deviation (SD) or more below the mean for young Caucasian adult women (T-score diagnostic criteria of -2.5), based on data that this criterion identified 30% of all postmenopausal women as having osteoporosis, more than half of whom would

Therefore, aBMD by DXA is not a suitable screening method for fracture risk in terms of accessibleness and cost. In addition, the correlations between bone strength and aBMD by DXA are reported to be 0.51-0.80 [8-11], which indicates aBMD only accounts for 50 to 80% of bone strength. And the application of aBMD measurements in isolation cannot identify individuals who eventually experience bone fracture because of the low sensitivity of the test [12].

Recently, quantitative ultrasound (QUS) is emerging as a relatively low-cost and readily accessible alternative means to identify osteoporosis, evaluate fracture risk, and initiate osteoporosis treatment. More recently, finite element (FE) method based on data from computed tomography (CT) has been used to assess bone strength, fracture risk, and therapeutic effects on osteoporosis.

#### **2. Dual energy X-ray absorptiometry (DXA)**

In the 1960s, a new method of measuring aBMD, called single-photon absorptiometry (SPA), was developed. In this method, a single-energy photon beam is passed through bone and soft tissue to a detector. The amount of mineral in the path is then quantified. This method most commonly uses a gamma-ray source coupled with a scintillation detector, which together scan across the area of interest [13]. The amount of the bone mineral in the tissue traversed by a well collimated gamma-ray beam is derived from its attenuation through bone plus soft tissue relative to that through soft tissue alone. The overall thickness of the soft tissue is standardized, usually by immersing the limb in water or cuffing with a fluidfilled bag. The value obtained is proportional to the bone mineral content of the segment scanned. The value may be divided by the bone width (yielding a result in g/cm) or by an estimate of the cross-sectional area to give a value for bone mineral density in g/cm2. The technique has been applied to the femur, humerus, metacarpal, os calcis, hand and foot, but the most commonly used site is the forearm. The most frequently used source is 125I (27keV), but has the major drawback of a relatively short half-life (60 days).

Accuracy may be compromised by a non-uniform thickness of fat, which has attenuation characteristics different from those of water or lean soft tissue. In some equipment, the program assumes the fat to be a uniform shell around the bone and makes a correction, but the correction requires a number of assumptions that influence the accuracy of the method. The heterogeneity of surrounding tissues is nevertheless considerably less than that of tissue surrounding axial sites such as the spine. Although true *in vivo* estimates of accuracy have not been made, errors in cadaveric studies of excised bone have sufficiently low to make the technique attractive for screening [14].

The radiation dose of SPA is very low and applied to a small volume of tissue, giving an effective dose equivalent of < 1µSv. Typical scanning times are 10-15 minutes. Single-energy X-ray absorptiometry (SXA) is a newly developed technique suitable for scanning appendicular sites. It avoids the need for isotopes and is likely to replace SPA.

The proximal femur and the vertebral bodies, with their associated processes, are very irregular bones that are difficult to delineate. Furthermore, they are surrounded by a widely varying amount of fat and muscle mass. The ratio of bone mass to soft tissue is thus lower in the spine or hip than in the forearm, and standardization of soft tissue by immersion in water is not feasible for these sites. These and other factors limit the use of SPA or SXA to

Therefore, aBMD by DXA is not a suitable screening method for fracture risk in terms of accessibleness and cost. In addition, the correlations between bone strength and aBMD by DXA are reported to be 0.51-0.80 [8-11], which indicates aBMD only accounts for 50 to 80% of bone strength. And the application of aBMD measurements in isolation cannot identify individuals who eventually experience bone fracture because of the low sensitivity of the

Recently, quantitative ultrasound (QUS) is emerging as a relatively low-cost and readily accessible alternative means to identify osteoporosis, evaluate fracture risk, and initiate osteoporosis treatment. More recently, finite element (FE) method based on data from computed tomography (CT) has been used to assess bone strength, fracture risk, and

In the 1960s, a new method of measuring aBMD, called single-photon absorptiometry (SPA), was developed. In this method, a single-energy photon beam is passed through bone and soft tissue to a detector. The amount of mineral in the path is then quantified. This method most commonly uses a gamma-ray source coupled with a scintillation detector, which together scan across the area of interest [13]. The amount of the bone mineral in the tissue traversed by a well collimated gamma-ray beam is derived from its attenuation through bone plus soft tissue relative to that through soft tissue alone. The overall thickness of the soft tissue is standardized, usually by immersing the limb in water or cuffing with a fluidfilled bag. The value obtained is proportional to the bone mineral content of the segment scanned. The value may be divided by the bone width (yielding a result in g/cm) or by an estimate of the cross-sectional area to give a value for bone mineral density in g/cm2. The technique has been applied to the femur, humerus, metacarpal, os calcis, hand and foot, but the most commonly used site is the forearm. The most frequently used source is 125I (27keV),

Accuracy may be compromised by a non-uniform thickness of fat, which has attenuation characteristics different from those of water or lean soft tissue. In some equipment, the program assumes the fat to be a uniform shell around the bone and makes a correction, but the correction requires a number of assumptions that influence the accuracy of the method. The heterogeneity of surrounding tissues is nevertheless considerably less than that of tissue surrounding axial sites such as the spine. Although true *in vivo* estimates of accuracy have not been made, errors in cadaveric studies of excised bone have sufficiently low to make the

The radiation dose of SPA is very low and applied to a small volume of tissue, giving an effective dose equivalent of < 1µSv. Typical scanning times are 10-15 minutes. Single-energy X-ray absorptiometry (SXA) is a newly developed technique suitable for scanning

The proximal femur and the vertebral bodies, with their associated processes, are very irregular bones that are difficult to delineate. Furthermore, they are surrounded by a widely varying amount of fat and muscle mass. The ratio of bone mass to soft tissue is thus lower in the spine or hip than in the forearm, and standardization of soft tissue by immersion in water is not feasible for these sites. These and other factors limit the use of SPA or SXA to

appendicular sites. It avoids the need for isotopes and is likely to replace SPA.

test [12].

therapeutic effects on osteoporosis.

technique attractive for screening [14].

**2. Dual energy X-ray absorptiometry (DXA)** 

but has the major drawback of a relatively short half-life (60 days).

the appendicular skeleton. The development of dual-photon absorptiometry (DPA) and, more recently, dual-energy X-ray absorptiometry (DXA) have resolved at least some of these problems. The different thickness of soft tissue can be accommodated by simultaneous measurement of the transmission of gamma-rays of two different energies, which makes the techniques applicable to any part of the body, but particularly the lumbar spine and hip.

The theory underlying DPA and DXA requires that there are only two components present – bone and soft tissue of uniform composition. In practice, fat forms a further component with attenuation characteristics that differ from those of water, muscle and most organs. A uniform layer of fat is unimportant, but fat is distributed non-uniformly in the region of the lumbar spine and may cause errors of up to 10% in spinal bone mineral. Errors can also be introduced by fat within the vertebral bone marrow.

Total body bone mineral can be measured by DPA, but instrumental problems are greater because of the wide range of count rates and the non-uniform distribution of fat, which introduces errors. However, total body bone mineral measured by neutron activation analysis. As with SPA, the radiation dose for DPA is low, the effective dose equivalent for part-body examinations being only a few microsieverts (µSv).

Recently, sources of gamma radiation have been replaced by X-ray generators. The necessary pairs of effective energies can be obtained either by K-edge filtering, using cerium or samarium, or by rapidly switching the generator potential. The advantages of these approaches are a higher beam intensity and therefore faster scan, improved spatial resolution with easier identification of vertebrae, and better precision. The absence of source decay also eliminates problems associated with decreasing count rates over the lifetime of the source.

Like DPA, DXA determines bone mineral density from an anterior-posterior image. The sites most commonly measured are the lumbar spine, generally L2-L4, including the intervertebral discs. Other sites include the hip, forearm, whole body and skeletal segments. The error in reproducibility *in vitro* is 1-2%. DXA has been reported to have a high shortterm and long-term precision *in vivo*, which is about twice that of DPA. This has led to its widespread use in studies of osteoporosis.

A recent development has been scanning of the lumbar spine in the lateral position, which has the advantage of eliminating the posterior arch and the spines of the vertebrae as well as aortic calcification from the measurement. Its limitations are the increased soft tissue mass and overlap of the projected image by the ribs and pelvis, so that only one or two vertebrae are measured. Lateral scanning provides a measurement of vertebral depth which, together with the antero-posterior area, can provide a volumetric measurement for calculating bone mineral mass per unit volume. Whether this volumetric density measure is a better predictor of fracture is unknown. The technique may be useful in assessment of bone density in children, allowing accurate assessment of vertebral size. The precision error of measurement of the vertebral body and mid-slice *in vivo* is of the order of 2% [15]. DXA has now largely replaced DPA for screening because of its greater precision, ease of use and freedom from several technical artifacts. The WHO defines osteoporosis as a value for aBMD by DXA 2.5 standard deviation (SD) or more below the mean for young Caucasian adult women (T-score diagnostic criteria of -2.5), based on data that this criterion identified 30% of all postmenopausal women as having osteoporosis, more than half of whom would have sustained a prior fracture [7].

*Ex Vivo* and *In Vivo* Assessment of Vertebral Strength

evaluating bone fracture risk.

and Vertebral Fracture Risk Assessed by Dual Energy X-Ray Absorptiometry 109

QUS bone assessment method has been recently introduced as an alternative for peripheral bone mass assessment, reflecting bone strength, bone density, and bone elasticity or fragility, and may be superior to aBMD by DXA [26]. The advantages of this method over Xray-based techniques, which include low cost, portability, and no radiation exposure, have encouraged the use of this method for defining a stage of development of osteoporosis and

There are several reports for assessing bone conditions *in vivo* using QUS method and apparatus. QUS devices can be classified mostly into 3 groups, related to the type of ultrasound transmission. Trabecular sound transmission is best for measuring the heel [27]. Cortical transverse transmission currently only is used in phalanx contact devices [28]. And cortical axial transmission presently is being investigated for use in phalanges, the radius, and the tibia [28]. Heel devices currently appear to have the most clinical applications, where QUS are being used and evaluated for the prediction of fracture risk, the diagnosis of osteoporosis, the initiation of osteoporosis treatment, the monitoring of osteoporosis treatment, and osteoporosis case finding. For these purposes, the recommended parameter of interest in clinical routine is a composite score, e.g., heel stiffness index or Quantitative Ultrasound Index (QUI) combining the results of broad-band ultrasound attenuation (BUA)

At the present time, there is good evidence that QUS can discriminate those with osteoporotic fractures from age-matched controls without osteoporotic fracture [29,30]. The power of heel QUS to predict fracture observed in cross-sectional studies has been confirmed prospectively in some populations as defined by sex, age, and ethnic background. This is particularly true of heel QUS and for hip and spinal fractures. However, because of methodological issues, it is difficult to compare studies. Nonetheless, it is possible to make the following generations. Using QUS of the heel, the increase in relative risk for each standard deviation decrease in stiffness index (SI) is approximately 2.0 for the hip and spine

The evidence from studies is good that the heel QUS SI using QUS devices is predictive of hip fracture risk in Caucasian and Asian women over age 55 and of any fracture risk in Asian women over age 55. Cortical axial transmission devices have no prospectively proven clinical utility, although clinical use in adults of phalanx QUS devices using cortical transverse transmission is also limited. These results for heel QUS are roughly the same as for DXA by BMD in terms of hip and spine fracture risk per SD decrease [12,42]. Discordant results between heel QUS and DXA, which are not infrequent, are not necessarily an indication of methodological error but rather due to the independence between the 2

Diagnosing osteoporosis using QUS is less supported by evidence and more complicated and problematic than assessing fracture risk. To start with, the T-score diagnostic criteria of - 2.5, classically used for DXA aBMD, cannot be applied to QUS without discrepancies in the numbers of women diagnosed with osteoporosis because of tremendous variations in QUS measurements by skeletal site, because different QUS devices yield different results, and because of the relatively poor correlation between heel QUS and hip/spine DXA measurements. If the prevalence of osteoporosis is defined as -2.5 SD from the mean

**4. Quantitative ultrasound (QUS) bone assessment method** 

and speed of sound (SOS), as measured in meters per second.

and roughly 1.5 for all fractures combined [31-41].

techniques.

#### **3. Quantitative computed tomography (QCT)**

In quantitative computed tomography (QCT), a thin transverse slice through the body is imaged. Under appropriate conditions, the image can be quantified to give a measure of volumetric bone mineral density (vBMD) (mg/cm3), and cancellous bone can be measured independently of surrounding cortical bone and aortic calcification. Developments have been concentrated in two directions: the construction of special equipment using a radionuclide source for measurements of the forearm, and the adaptation of X-ray CT machines installed for general radiology to measure vBMD. The attraction of the technique is that cancellous bone can be examined separately from cortical bone. It also gives a true value for mineral density (mg/cm3) unlike other techniques.

A dedicated forearm scanner was first described in the mid 1970s [16,17]. The photon source is 125I and is mounted in a gantry with a sodium iodide scintillation detector. A linear scan is performed at each of 48 angular positions. Computer reconstruction generates an image in which a region of interest in the cancellous bone of the distal ulna is selected. Since 1980s, QCT has been used as a means for non-invasive quantitative determination of bone mineral of the spine [18,19].

A lateral plane projection scan is necessary for precise slice positioning through the centers of the vertebrae. Comparison between the CT Hounsfield numbers and a calibration standard scanned simultaneously allows bone density to be expressed in terms of the equivalent concentration of the material of the standard. Regions of interest within the vertebral bodies are selected: circular, elliptical, rectangular or other chosen areas are selected to include all the cancellous bone just inside the cortex. The relationship between the observed CT number and the true attenuation coefficient is subject to short- and longterm variation, so that it is necessary to scan the patient and a calibration standard simultaneously. Recently, simple standards with fewer components based on suspensions of calcium hydroxyapatite in plastic have been adopted. Comparison between the standard and the Hounsfield numbers of the trabecular region of the vertebral bodies allows bone density to be expressed in terms of the equivalent concentration of the material of the standard.

Investigators reported the prediction of vertebral body compressive strength using QCT. In 1985, McBroom et al. [20] showed a strong positive correlation between QCT and apparent density of the vertebral trabecular bone but could find only suggestive, not quite significant, correlations between QCT and the vertebral body compressive strength. Cann et al. [21] showed that QCT evaluation of vertebral trabecular bone mineral density is a useful tool for determining the patients with increased risk of vertebral fracture. The positive correlations between QCT and vertebral body compressive strength in cadaver studies are 0.72-0.74 [22,23].

The biggest source of error in X-ray CT systems is fat within the bone marrow: accuracy errors of up to 30%. The accuracy can be improved by carrying out scans at two different potentials (dual energy techniques); typically, 80 and 120 kVp are used. Kalender et al. [24] claim an accuracy error of 5% *in vitro*, but errors *in vivo* are likely to be larger. The effective radiation doses equivalent for QCT are 0.3 mSv for single energy techniques and 1 mSv for dual energy techniques, respectively [25].

In quantitative computed tomography (QCT), a thin transverse slice through the body is imaged. Under appropriate conditions, the image can be quantified to give a measure of volumetric bone mineral density (vBMD) (mg/cm3), and cancellous bone can be measured independently of surrounding cortical bone and aortic calcification. Developments have been concentrated in two directions: the construction of special equipment using a radionuclide source for measurements of the forearm, and the adaptation of X-ray CT machines installed for general radiology to measure vBMD. The attraction of the technique is that cancellous bone can be examined separately from cortical bone. It also gives a true

A dedicated forearm scanner was first described in the mid 1970s [16,17]. The photon source is 125I and is mounted in a gantry with a sodium iodide scintillation detector. A linear scan is performed at each of 48 angular positions. Computer reconstruction generates an image in which a region of interest in the cancellous bone of the distal ulna is selected. Since 1980s, QCT has been used as a means for non-invasive quantitative determination of bone mineral

A lateral plane projection scan is necessary for precise slice positioning through the centers of the vertebrae. Comparison between the CT Hounsfield numbers and a calibration standard scanned simultaneously allows bone density to be expressed in terms of the equivalent concentration of the material of the standard. Regions of interest within the vertebral bodies are selected: circular, elliptical, rectangular or other chosen areas are selected to include all the cancellous bone just inside the cortex. The relationship between the observed CT number and the true attenuation coefficient is subject to short- and longterm variation, so that it is necessary to scan the patient and a calibration standard simultaneously. Recently, simple standards with fewer components based on suspensions of calcium hydroxyapatite in plastic have been adopted. Comparison between the standard and the Hounsfield numbers of the trabecular region of the vertebral bodies allows bone density to be expressed in terms of the equivalent concentration of the material of the

Investigators reported the prediction of vertebral body compressive strength using QCT. In 1985, McBroom et al. [20] showed a strong positive correlation between QCT and apparent density of the vertebral trabecular bone but could find only suggestive, not quite significant, correlations between QCT and the vertebral body compressive strength. Cann et al. [21] showed that QCT evaluation of vertebral trabecular bone mineral density is a useful tool for determining the patients with increased risk of vertebral fracture. The positive correlations between QCT and vertebral body compressive strength in cadaver studies are 0.72-0.74

The biggest source of error in X-ray CT systems is fat within the bone marrow: accuracy errors of up to 30%. The accuracy can be improved by carrying out scans at two different potentials (dual energy techniques); typically, 80 and 120 kVp are used. Kalender et al. [24] claim an accuracy error of 5% *in vitro*, but errors *in vivo* are likely to be larger. The effective radiation doses equivalent for QCT are 0.3 mSv for single energy techniques and 1 mSv for

**3. Quantitative computed tomography (QCT)** 

value for mineral density (mg/cm3) unlike other techniques.

of the spine [18,19].

standard.

[22,23].

dual energy techniques, respectively [25].

#### **4. Quantitative ultrasound (QUS) bone assessment method**

QUS bone assessment method has been recently introduced as an alternative for peripheral bone mass assessment, reflecting bone strength, bone density, and bone elasticity or fragility, and may be superior to aBMD by DXA [26]. The advantages of this method over Xray-based techniques, which include low cost, portability, and no radiation exposure, have encouraged the use of this method for defining a stage of development of osteoporosis and evaluating bone fracture risk.

There are several reports for assessing bone conditions *in vivo* using QUS method and apparatus. QUS devices can be classified mostly into 3 groups, related to the type of ultrasound transmission. Trabecular sound transmission is best for measuring the heel [27]. Cortical transverse transmission currently only is used in phalanx contact devices [28]. And cortical axial transmission presently is being investigated for use in phalanges, the radius, and the tibia [28]. Heel devices currently appear to have the most clinical applications, where QUS are being used and evaluated for the prediction of fracture risk, the diagnosis of osteoporosis, the initiation of osteoporosis treatment, the monitoring of osteoporosis treatment, and osteoporosis case finding. For these purposes, the recommended parameter of interest in clinical routine is a composite score, e.g., heel stiffness index or Quantitative Ultrasound Index (QUI) combining the results of broad-band ultrasound attenuation (BUA) and speed of sound (SOS), as measured in meters per second.

At the present time, there is good evidence that QUS can discriminate those with osteoporotic fractures from age-matched controls without osteoporotic fracture [29,30]. The power of heel QUS to predict fracture observed in cross-sectional studies has been confirmed prospectively in some populations as defined by sex, age, and ethnic background. This is particularly true of heel QUS and for hip and spinal fractures. However, because of methodological issues, it is difficult to compare studies. Nonetheless, it is possible to make the following generations. Using QUS of the heel, the increase in relative risk for each standard deviation decrease in stiffness index (SI) is approximately 2.0 for the hip and spine and roughly 1.5 for all fractures combined [31-41].

The evidence from studies is good that the heel QUS SI using QUS devices is predictive of hip fracture risk in Caucasian and Asian women over age 55 and of any fracture risk in Asian women over age 55. Cortical axial transmission devices have no prospectively proven clinical utility, although clinical use in adults of phalanx QUS devices using cortical transverse transmission is also limited. These results for heel QUS are roughly the same as for DXA by BMD in terms of hip and spine fracture risk per SD decrease [12,42]. Discordant results between heel QUS and DXA, which are not infrequent, are not necessarily an indication of methodological error but rather due to the independence between the 2 techniques.

Diagnosing osteoporosis using QUS is less supported by evidence and more complicated and problematic than assessing fracture risk. To start with, the T-score diagnostic criteria of - 2.5, classically used for DXA aBMD, cannot be applied to QUS without discrepancies in the numbers of women diagnosed with osteoporosis because of tremendous variations in QUS measurements by skeletal site, because different QUS devices yield different results, and because of the relatively poor correlation between heel QUS and hip/spine DXA measurements. If the prevalence of osteoporosis is defined as -2.5 SD from the mean

*Ex Vivo* and *In Vivo* Assessment of Vertebral Strength

yield strength determined.

and Vertebral Fracture Risk Assessed by Dual Energy X-Ray Absorptiometry 111

bone mineral density, this distribution is used to define bone material properties, and the FE method of analysis is used to determine structural properties of the whole or a part of the bone. This information is then used to predict risk of fracture under specified loading conditions. Specifically, the distribution of bone material properties determined noninvasively is used as input to a FE analysis of structural strength, and other parameters such as loading conditions and boundary conditions are also included in the model as needed. Using mathematical methods contained in commercially-available or specially written computer programs, the model of a bone can be incrementally loaded until failure, and the

A FE method based on data from CT has been applied to predict proximal femoral fracture [66-70]. CT-based FE method appears more predictive of femoral strength than QCT or DXA alone [66] and can predict proximal femoral fracture location [68]. Nonlinear FE method demonstrated improved predictions of femoral strength [69]. For the spine, CT-based nonlinear FE method was clinically applied to assess vertebral strength [71] and cadaver studies have been performed to evaluate the accuracy of CT-based FE method [72-77]. The cadaver studies have verified CT-based FE method predicts failure loads and fracture patterns for 10-mm-thick vertebral sections [72] and can predict *ex vivo* vertebral compressive strength better than aBMD [73,74] and QCT alone [75]. CT-based nonlinear FE method can accurately predict vertebral strength, fracture sites and distribution of minimum principal strain *ex vivo* [77]. Based on verification by the cadaver studies, FE method has been applied clinically to the assessment of chronic glucocorticoid treatment at the hip [78], as well as teriparatide and alendronate treatment for osteoporosis at the lumbar spine [79],

A study assessing vertebral fracture risk and medication effects on osteoporosis *in vivo* with CT-based nonlinear FE method showed that analyzed vertebral compressive strength had stronger discriminatory power for vertebral fracture than aBMD and vBMD, and detected alendronate effects at 3 months earlier than aBMD and vBMD [80]. The CV (coefficient of variation) for the measurement of vertebral compressive strength was 0.96% *ex vivo*. The

CT-based FE method predicts compressive bone strength accurately and is useful for assessing the risk of fracture and therapeutic effects on osteoporosis, and provides unique theories from a biomechanical perspective. This method also predicts bone strength under specified loading conditions such as those normally seen in activities of daily living [81,82].

This study was conducted at Tokyo University in Tokyo, Japan. The study protocol was

Twelve thoracolumbar (T11, T12, and L1) vertebrae with no skeletal pathologies were collected within 24 hours of death from 4 males (31, 55, 67, and 83 years old). Causes of death for the four donors were myelodysplastic syndrome, pneumonia, adult T-cell leukemia, and bladder cancer, respectively. All of the specimens were obtained at Tokyo University Hospital with the approval of the ethics committee and with informed consent. They were stored at –70 C after each step in the protocol. The vertebrae were disarticulated,

proving useful for assessing medication effects on bone strength.

**6. Assessment of vertebral strength** *ex vivo* **by DXA** 

approved by the ethics committee.

effective radiation dose for assessing vertebral compressive strength is 3 mSv.

threshold for QUS, even within the same sample population, different QUS instruments and different skeletal sites generate prevalence estimates that vary as much as 10-fold, such as prevalence estimates among Caucasian women over age 65 ranging from 4 to 50% [43-46]. To overcome this dilemma, there is a need for predefined, device-specific diagnostic thresholds. One recommended system suggests calibrating QUS measurements with DXA results, the latter used as the "gold standard," so that an upper QUS threshold is set to identify osteoporosis with 90% sensitivity and a lower threshold is set to identify osteoporosis with 90% specificity [47]. Using such a system, one could identify osteoporosis with high probability in patients whose results fall below the lower threshold for QUS, where specificity exceeds 90%; between the upper and lower thresholds, the diagnosis of osteoporosis would be considered quite equivocal, so that another means of measurement, like DXA aBMD, would be highly recommended; and above the upper threshold for QUS, where the sensitivity of a value below the threshold is 90%, osteoporosis would be deemed unlikely.

Except in the case of a low-energy fractures of the hip or spine, when the fracture alone is adequate to require treatment, all currently published recommendations for the initiation of treatment for osteoporosis are based on DXA aBMD values; in no instance, to date, are the results of QUS the definitive parameter. Despite this, several studies have demonstrated high levels of correlation between heel trabecular sound transmission and aBMD at matched skeletal sites [48-50]. Moreover, both SOS and BUA, standard QUS measurements, are dependent on overall bone strength which, in turn, is related to bone density, architecture and turnover, and the extent of bone mineralization [48,50,51-56]. These factors likely work together to maintain the overall quality and strength of bone and to prevent fractures and other bone failure. QUS parameters of heel trabecular transverse transmission are highly correlated with bone strength [57-62]. Consequently, it is conceivable that QUS guidelines for treatment initiation could be created, especially if combined with the use of clinical risk factors [63]. But no randomized clinical trials have been published examining whether individuals identified as high risk for fracture by QUS respond to treatment.

#### **5. Finite element (FE) method based on data from computed tomography**

The finite element (FE) method, an advanced computer technique of structural stress analysis developed in engineering mechanics, was first introduced to orthopaedic biomechanics in 1972 to evaluate stressed in human bones [64]. Since then, this method has been used to study the mechanics of human bones [65]. In the early 1990s, the FE method of analyzing a bone for fracture risk using 3-dimensional CT data was developed.

The object of this method is to measure non-invasively the strength of an individual bone in an individual patient. This measurement can then be used to determine whether or not the bone will fracture under specified loading conditions such as those normally seen in daily living. It can also be used to estimate fracture risks under abnormal loading conditions such as occur in falling, jumping or during athletic events or heavy training regimens. This method uses the distribution of physical properties of bone measured non-invasively in an individual and mathematical analysis of that distribution to predict the risk that a bone may fracture under applied loads. The use of such methods relates to the clinical disease of osteoporosis, or in general metabolic bone diseases. In a primary application, 3-dimensional CT data acquired using a conventional CT scanner are used to determine the distribution of

threshold for QUS, even within the same sample population, different QUS instruments and different skeletal sites generate prevalence estimates that vary as much as 10-fold, such as prevalence estimates among Caucasian women over age 65 ranging from 4 to 50% [43-46]. To overcome this dilemma, there is a need for predefined, device-specific diagnostic thresholds. One recommended system suggests calibrating QUS measurements with DXA results, the latter used as the "gold standard," so that an upper QUS threshold is set to identify osteoporosis with 90% sensitivity and a lower threshold is set to identify osteoporosis with 90% specificity [47]. Using such a system, one could identify osteoporosis with high probability in patients whose results fall below the lower threshold for QUS, where specificity exceeds 90%; between the upper and lower thresholds, the diagnosis of osteoporosis would be considered quite equivocal, so that another means of measurement, like DXA aBMD, would be highly recommended; and above the upper threshold for QUS, where the sensitivity of a value below the threshold is 90%, osteoporosis would be deemed

Except in the case of a low-energy fractures of the hip or spine, when the fracture alone is adequate to require treatment, all currently published recommendations for the initiation of treatment for osteoporosis are based on DXA aBMD values; in no instance, to date, are the results of QUS the definitive parameter. Despite this, several studies have demonstrated high levels of correlation between heel trabecular sound transmission and aBMD at matched skeletal sites [48-50]. Moreover, both SOS and BUA, standard QUS measurements, are dependent on overall bone strength which, in turn, is related to bone density, architecture and turnover, and the extent of bone mineralization [48,50,51-56]. These factors likely work together to maintain the overall quality and strength of bone and to prevent fractures and other bone failure. QUS parameters of heel trabecular transverse transmission are highly correlated with bone strength [57-62]. Consequently, it is conceivable that QUS guidelines for treatment initiation could be created, especially if combined with the use of clinical risk factors [63]. But no randomized clinical trials have been published examining whether

individuals identified as high risk for fracture by QUS respond to treatment.

analyzing a bone for fracture risk using 3-dimensional CT data was developed.

**5. Finite element (FE) method based on data from computed tomography** 

The finite element (FE) method, an advanced computer technique of structural stress analysis developed in engineering mechanics, was first introduced to orthopaedic biomechanics in 1972 to evaluate stressed in human bones [64]. Since then, this method has been used to study the mechanics of human bones [65]. In the early 1990s, the FE method of

The object of this method is to measure non-invasively the strength of an individual bone in an individual patient. This measurement can then be used to determine whether or not the bone will fracture under specified loading conditions such as those normally seen in daily living. It can also be used to estimate fracture risks under abnormal loading conditions such as occur in falling, jumping or during athletic events or heavy training regimens. This method uses the distribution of physical properties of bone measured non-invasively in an individual and mathematical analysis of that distribution to predict the risk that a bone may fracture under applied loads. The use of such methods relates to the clinical disease of osteoporosis, or in general metabolic bone diseases. In a primary application, 3-dimensional CT data acquired using a conventional CT scanner are used to determine the distribution of

unlikely.

bone mineral density, this distribution is used to define bone material properties, and the FE method of analysis is used to determine structural properties of the whole or a part of the bone. This information is then used to predict risk of fracture under specified loading conditions. Specifically, the distribution of bone material properties determined noninvasively is used as input to a FE analysis of structural strength, and other parameters such as loading conditions and boundary conditions are also included in the model as needed. Using mathematical methods contained in commercially-available or specially written computer programs, the model of a bone can be incrementally loaded until failure, and the yield strength determined.

A FE method based on data from CT has been applied to predict proximal femoral fracture [66-70]. CT-based FE method appears more predictive of femoral strength than QCT or DXA alone [66] and can predict proximal femoral fracture location [68]. Nonlinear FE method demonstrated improved predictions of femoral strength [69]. For the spine, CT-based nonlinear FE method was clinically applied to assess vertebral strength [71] and cadaver studies have been performed to evaluate the accuracy of CT-based FE method [72-77]. The cadaver studies have verified CT-based FE method predicts failure loads and fracture patterns for 10-mm-thick vertebral sections [72] and can predict *ex vivo* vertebral compressive strength better than aBMD [73,74] and QCT alone [75]. CT-based nonlinear FE method can accurately predict vertebral strength, fracture sites and distribution of minimum principal strain *ex vivo* [77]. Based on verification by the cadaver studies, FE method has been applied clinically to the assessment of chronic glucocorticoid treatment at the hip [78], as well as teriparatide and alendronate treatment for osteoporosis at the lumbar spine [79], proving useful for assessing medication effects on bone strength.

A study assessing vertebral fracture risk and medication effects on osteoporosis *in vivo* with CT-based nonlinear FE method showed that analyzed vertebral compressive strength had stronger discriminatory power for vertebral fracture than aBMD and vBMD, and detected alendronate effects at 3 months earlier than aBMD and vBMD [80]. The CV (coefficient of variation) for the measurement of vertebral compressive strength was 0.96% *ex vivo*. The effective radiation dose for assessing vertebral compressive strength is 3 mSv.

CT-based FE method predicts compressive bone strength accurately and is useful for assessing the risk of fracture and therapeutic effects on osteoporosis, and provides unique theories from a biomechanical perspective. This method also predicts bone strength under specified loading conditions such as those normally seen in activities of daily living [81,82].

#### **6. Assessment of vertebral strength** *ex vivo* **by DXA**

This study was conducted at Tokyo University in Tokyo, Japan. The study protocol was approved by the ethics committee.

Twelve thoracolumbar (T11, T12, and L1) vertebrae with no skeletal pathologies were collected within 24 hours of death from 4 males (31, 55, 67, and 83 years old). Causes of death for the four donors were myelodysplastic syndrome, pneumonia, adult T-cell leukemia, and bladder cancer, respectively. All of the specimens were obtained at Tokyo University Hospital with the approval of the ethics committee and with informed consent. They were stored at –70 C after each step in the protocol. The vertebrae were disarticulated,

*Ex Vivo* and *In Vivo* Assessment of Vertebral Strength

(DPX; Lunar, Madison, WI, USA).

**8. Discussion** 

**7. Assessment of vertebral fracture risk** *in vivo* **by DXA** 

Windows version 5.0 software (SAS Institute, Cary, NC, USA).

0.759 ± 0.207 g/cm2 (Mann-Whitney *U* test, *p* = 0.0255).

written informed consent in accordance with the Declaration of Helsinki.

and Vertebral Fracture Risk Assessed by Dual Energy X-Ray Absorptiometry 113

This study was conducted at Tokyo Metropolitan Geriatric Medical Center in Tokyo, Japan. The study protocol was approved by the ethics committee and each participant provided

The inclusion criteria included ambulatory postmenopausal Japanese women aged between 49 and 85 years old. Exclusion criteria included women with any disorders of bone and mineral metabolism other than postmenopausal osteoporosis, those who had any recent or current treatment with the potential to alter bone turnover or bone metabolism. Vertebral fracture was diagnosed based on lateral spine radiography. Radiographic vertebral fracture was defined if either the anterior or central height was 20% less than posterior height. A total of 123 eligible participants were enrolled in this cross-sectional study. For all participants, aBMD of the anteroposterior (AP) lumbar spine (L2-4) were measured by DXA

Logistic regression analysis was performed to estimate risk factors for vertebral fracture. L2- 4 aBMD was assessed using sensitivity and specificity curves to determine the optimal cutoff point as the vertebral fracture threshold. For each statistical analysis, differences were considered significant at *p*<0.05. Statistical analysis was performed using StatView for

The 123 women enrolled in the *in vivo* clinical study had a mean age of 71.8 ±7.4 years, mean height of 149.4 ±5.6 cm, and mean weight of 50.2 ±7.4 kg. Measured L2-4 aBMD was 0.816 ±0.191 g/cm2. Subjects were classified on the basis of prior vertebral fracture. Among the 123 women, 75 subjects did not have any vertebral fractures (nonfracture group) and 48 subjects already had vertebral fractures (fracture group). The average aBMD of the nonfracture group was 0.860 ± 0.166 g/cm2, which was greater than that of the fracture group at

Vertebral fractures were present in 39.0% of the total study population. Among the fracture group, vertebral fractures spontaneously developed in 29 women (spontaneous fracture group) and were caused by trauma in 19 women (traumatic fracture group). Among the 19 subjects in the traumatic fracture group, 18 women developed fracture following a fall from standing height, and 1 woman developed fracture following a fall down stairs. To exclude factors of trauma, 75 subjects in the nonfracture group and 29 subjects in the spontaneous fracture group were compared. aBMD (Mann-Whitney *U* test, *p*=0.0033) was significantly decreased in the spontaneous fracture group compared with the nonfracture group. Logistic regression analysis after adjustment for age and body weight revealed that aBMD reduction as risk factors associated with spontaneous vertebral fracture, the odds ratio per SD decrease was 1.83 with 1.13-3.26 of 95% confidence interval (*p*=0.0238). aBMD was also assessed by sensitivity and specificity curves. The nonfracture group and spontaneous fracture group (104 women in total) were assessed in a cross-sectional manner. The optimal point on the sensitivity and specificity curves used as the fracture threshold to predict spontaneous vertebral fractures

for aBMD was 0.816 g/cm2 with 69.0% sensitivity and 72.0% specificity (Fig. 2).

Bone strength primarily reflects bone density and bone quality, which are influenced by bone architecture, turnover, accumulation of damage, and mineralization [5]. Previous

and the discs were excised. Then the posterior elements of each vertebra were removed by cutting through the pedicles. The vertebrae were immersed in water and aBMD (g/cm2) of the vertebrae were measured by DXA (DPX; Lunar, Madison, WI, USA) in the supine position.

To assess vertebral strength, a quasi-static uniaxial compression test of each vertebra was conducted. To restrain the specimens for load testing, both upper and lower surfaces of the vertebrae were embedded in dental resin (Ostron; GC Dental Products Co., Aichi, Japan) so that the two surfaces were exactly parallel. Then the embedded specimens were placed on a mechanical testing machine (TENSILON UTM-2.5T; Orientec, Tokyo, Japan) and were compressed at a cross-head displacement rate of 0.5 mm per minute. A compression plate with a ball joint was used to apply a uniform load onto the upper surface of the specimen. The applied load was measured by a load cell (T-CLB-5-F-SR; T. S. Engineering, Kanagawa, Japan). The load was recorded using MacLab/4(AD Instruments, Castle Hill, NSW, Australia)at a sampling rate of 2 Hz. The measured vertebral strength was defined as the ultimate load achieved. Pearson's correlation analysis was used to evaluate correlations between the measured aBMD by DXA and the measured vertebral strength by mechanical testing.

The result from the *ex vivo* assessment, aBMD by DXA ranged from 0.287 to 0.705 g/cm2, while the measured vertebral strength by mechanical testing ranged from 1.54 to 4.62 kN. There were significant linear correlations between aBMD and the measured vertebral strength (r = 0.915, *p* < 0.0001) (Fig. 1).

Fig. 1. The experimentally measured vertebral strength versus aBMD measured by DXA. They were significantly correlated.

#### **7. Assessment of vertebral fracture risk** *in vivo* **by DXA**

112 Dual Energy X-Ray Absorptiometry

and the discs were excised. Then the posterior elements of each vertebra were removed by cutting through the pedicles. The vertebrae were immersed in water and aBMD (g/cm2) of the vertebrae were measured by DXA (DPX; Lunar, Madison, WI, USA) in the supine

To assess vertebral strength, a quasi-static uniaxial compression test of each vertebra was conducted. To restrain the specimens for load testing, both upper and lower surfaces of the vertebrae were embedded in dental resin (Ostron; GC Dental Products Co., Aichi, Japan) so that the two surfaces were exactly parallel. Then the embedded specimens were placed on a mechanical testing machine (TENSILON UTM-2.5T; Orientec, Tokyo, Japan) and were compressed at a cross-head displacement rate of 0.5 mm per minute. A compression plate with a ball joint was used to apply a uniform load onto the upper surface of the specimen. The applied load was measured by a load cell (T-CLB-5-F-SR; T. S. Engineering, Kanagawa, Japan). The load was recorded using MacLab/4(AD Instruments, Castle Hill, NSW, Australia)at a sampling rate of 2 Hz. The measured vertebral strength was defined as the ultimate load achieved. Pearson's correlation analysis was used to evaluate correlations between the measured aBMD by DXA and the measured vertebral strength by mechanical

The result from the *ex vivo* assessment, aBMD by DXA ranged from 0.287 to 0.705 g/cm2, while the measured vertebral strength by mechanical testing ranged from 1.54 to 4.62 kN. There were significant linear correlations between aBMD and the measured vertebral

Fig. 1. The experimentally measured vertebral strength versus aBMD measured by DXA.

position.

testing.

strength (r = 0.915, *p* < 0.0001) (Fig. 1).

They were significantly correlated.

This study was conducted at Tokyo Metropolitan Geriatric Medical Center in Tokyo, Japan. The study protocol was approved by the ethics committee and each participant provided written informed consent in accordance with the Declaration of Helsinki.

The inclusion criteria included ambulatory postmenopausal Japanese women aged between 49 and 85 years old. Exclusion criteria included women with any disorders of bone and mineral metabolism other than postmenopausal osteoporosis, those who had any recent or current treatment with the potential to alter bone turnover or bone metabolism. Vertebral fracture was diagnosed based on lateral spine radiography. Radiographic vertebral fracture was defined if either the anterior or central height was 20% less than posterior height. A total of 123 eligible participants were enrolled in this cross-sectional study. For all participants, aBMD of the anteroposterior (AP) lumbar spine (L2-4) were measured by DXA (DPX; Lunar, Madison, WI, USA).

Logistic regression analysis was performed to estimate risk factors for vertebral fracture. L2- 4 aBMD was assessed using sensitivity and specificity curves to determine the optimal cutoff point as the vertebral fracture threshold. For each statistical analysis, differences were considered significant at *p*<0.05. Statistical analysis was performed using StatView for Windows version 5.0 software (SAS Institute, Cary, NC, USA).

The 123 women enrolled in the *in vivo* clinical study had a mean age of 71.8 ±7.4 years, mean height of 149.4 ±5.6 cm, and mean weight of 50.2 ±7.4 kg. Measured L2-4 aBMD was 0.816 ±0.191 g/cm2. Subjects were classified on the basis of prior vertebral fracture. Among the 123 women, 75 subjects did not have any vertebral fractures (nonfracture group) and 48 subjects already had vertebral fractures (fracture group). The average aBMD of the nonfracture group was 0.860 ± 0.166 g/cm2, which was greater than that of the fracture group at 0.759 ± 0.207 g/cm2 (Mann-Whitney *U* test, *p* = 0.0255).

Vertebral fractures were present in 39.0% of the total study population. Among the fracture group, vertebral fractures spontaneously developed in 29 women (spontaneous fracture group) and were caused by trauma in 19 women (traumatic fracture group). Among the 19 subjects in the traumatic fracture group, 18 women developed fracture following a fall from standing height, and 1 woman developed fracture following a fall down stairs. To exclude factors of trauma, 75 subjects in the nonfracture group and 29 subjects in the spontaneous fracture group were compared. aBMD (Mann-Whitney *U* test, *p*=0.0033) was significantly decreased in the spontaneous fracture group compared with the nonfracture group. Logistic regression analysis after adjustment for age and body weight revealed that aBMD reduction as risk factors associated with spontaneous vertebral fracture, the odds ratio per SD decrease was 1.83 with 1.13-3.26 of 95% confidence interval (*p*=0.0238). aBMD was also assessed by sensitivity and specificity curves. The nonfracture group and spontaneous fracture group (104 women in total) were assessed in a cross-sectional manner. The optimal point on the sensitivity and specificity curves used as the fracture threshold to predict spontaneous vertebral fractures for aBMD was 0.816 g/cm2 with 69.0% sensitivity and 72.0% specificity (Fig. 2).

#### **8. Discussion**

Bone strength primarily reflects bone density and bone quality, which are influenced by bone architecture, turnover, accumulation of damage, and mineralization [5]. Previous

*Ex Vivo* and *In Vivo* Assessment of Vertebral Strength

micro-CT data with less radiation dose might be promising.

Disease. Osteoporos Int. 1997;7:390-406.

Organ Tech Rep Ser 1994;843:1-129.

Biomecha 1994; 9: 180-186.

1728.

osteoporotic fractures. Epidemiol Rev. 1985;7:178-208.

**9. References** 

1997;87:398-403.

and Vertebral Fracture Risk Assessed by Dual Energy X-Ray Absorptiometry 115

This method assesses bone geometry and heterogeneous bone mass distribution as well as aBMD, but cannot detect microdamage and bone turnover. In clinical application, other parameters such as age and bone turnover markers should be included to assess the risk of fracture and therapeutic effects. Methods for assessing fracture risk and therapeutic effects

Prediction by FE method with a smaller element size using the data from CT scans with a thinner slice thickness and a smaller pixel size is more accurate. On the other hand, thinner CT slices lead to more radiation exposure in the clinical situation. To decrease radiation exposure as much as possible during CT scanning, optimization of the element size of the FE method was performed by assessing the accuracy of the FE method simulation [83]. With the limited resolution of currently available CT scanners, the micro-architecture of the bone cannot be precisely assessed. Micro-CT and synchrotron micro-CT visualize bone microstructure. However, obtaining micro-CT scans of a whole vertebra *in vivo* would be impossible with the currently available scanners. Also, use of thinner CT slices to obtain images leads to more radiation exposure. With future developments, FE method based on

[1] Cummings SR, Kelsey JL, Nevitt MC, O'Dowd KJ. Epidemiology of osteoporosis and

[2] Eastell R, Reid DM, Compston J, *et al*. Secondary prevention of osteoporosis: when should a non-vertebral fracture be a trigger for action? Q J Med 2001;94:575-597. [3] Wolinsky FD, Fitzgerald JF, Stump TE. The effect of hip fracture on mortality,

[5] NIH Consensus Development Panel on Osteoporosis Prevention, Diagnosis, and

[6] Kanis JA, Delmas P, Burckhardt P, Cooper C, Torgerson D. Guidelines for diagnosis and

[7] World Health Organization. Assessment of fracture risk and its application to screening

[8] Edmondston SJ, Singer KP, Day RE, *et al.* In-vitro relationships between vertebral body

[9] Cheng XG, Nicholson PH, Boonen S, *et al.* Prediction of vertebral strength in vitro by

[10] Myers BS, Arbogast KB, Lobaugh B, Harper KD, Richardson WJ, Drezner MK.

energy x-ray absorptiometry. J Bone Miner Res 1994;9:687-693.

[4] Melton LJ 3rd. Epidemiology of spinal osteoporosis. Spine 1997;22(suppl):2-11.

hospitalization, and functional status: a prospective study. Am J Public Health

Therapy. Osteoporosis prevention, diagnosis, and therapy. JAMA 2001;285:785-795.

management of osteoporosis. The European Foundation for Osteoporosis and Bone

for postmenopausal osteoporosis. Report of a WHO Study Group. World Health

density, size and compressive strength in the elderly thoracolumbar spine. Clin

spinal bone densitometry and calcaneal ultrasound. J Bone Miner Res 1997;12:1721-

Improved assessment of lumbar vertebral body strength using supine lateral dual-

on osteoporosis in the future might include other parameters as well as CT data.

Fig. 2. Sensitivity and specificity curves to determine the optimal cut-off point of aBMD measured by DXA to predict spontaneous vertebral fracture.

studies showed that aBMD explained 50-80% of vertebral strength [8-11] based on data that the correlations between aBMD and the measured vertebral strength were 0.51 to 0.80. In this study, the correlations between the measured values of aBMD and the vertebral strength were 0.915 and better than the previous studies. This *ex vivo* study showed that aBMD measurements in isolation might assess vertebral strength well.

In the treatment of osteoporosis, the target is to assess fracture risk and prevent fractures. This *in vivo* study showed that aBMD had high discriminatory power for spontaneous vertebral fracture. The cut-off value of aBMD for predicting vertebral fractures without trauma was 0.816 g/cm2, equivalent to -2.62 SD compared to young healthy Japanese women. Low trauma fractures such as a fall from a standing height are due to osteoporosis. The present assessment excluded the traumatic fracture group. Therefore, the threshold value was not for diagnosing osteoporosis, but for assessing spontaneous vertebral fracture risk.

This *ex vivo* and *in vivo* study showed that aBMD was a good parameter of vertebral strength and vertebral fracture risk. However, aBMD in isolation can only assess bone density and can not assess bone quality. Therefore, accuracy of assessing vertebral strength by aBMD is limited.

To improve accuracy of assessing vertebral strength and vertebral fracture risk, new method has been developed. CT-based nonlinear FE method can accurately predict vertebral strength, fracture sites and distribution of minimum principal strain *ex vivo* [77]. Based on verification by the cadaver studies, FE method has been applied clinically. A study assessing vertebral fracture risk and medication effects on osteoporosis *in vivo* with CT-based nonlinear FE method showed that analyzed vertebral compressive strength had stronger discriminatory power for vertebral fracture than aBMD and detected alendronate effects at 3 months earlier than aBMD [80].

This method assesses bone geometry and heterogeneous bone mass distribution as well as aBMD, but cannot detect microdamage and bone turnover. In clinical application, other parameters such as age and bone turnover markers should be included to assess the risk of fracture and therapeutic effects. Methods for assessing fracture risk and therapeutic effects on osteoporosis in the future might include other parameters as well as CT data.

Prediction by FE method with a smaller element size using the data from CT scans with a thinner slice thickness and a smaller pixel size is more accurate. On the other hand, thinner CT slices lead to more radiation exposure in the clinical situation. To decrease radiation exposure as much as possible during CT scanning, optimization of the element size of the FE method was performed by assessing the accuracy of the FE method simulation [83]. With the limited resolution of currently available CT scanners, the micro-architecture of the bone cannot be precisely assessed. Micro-CT and synchrotron micro-CT visualize bone microstructure. However, obtaining micro-CT scans of a whole vertebra *in vivo* would be impossible with the currently available scanners. Also, use of thinner CT slices to obtain images leads to more radiation exposure. With future developments, FE method based on micro-CT data with less radiation dose might be promising.

#### **9. References**

114 Dual Energy X-Ray Absorptiometry

Fig. 2. Sensitivity and specificity curves to determine the optimal cut-off point of aBMD

studies showed that aBMD explained 50-80% of vertebral strength [8-11] based on data that the correlations between aBMD and the measured vertebral strength were 0.51 to 0.80. In this study, the correlations between the measured values of aBMD and the vertebral strength were 0.915 and better than the previous studies. This *ex vivo* study showed that

In the treatment of osteoporosis, the target is to assess fracture risk and prevent fractures. This *in vivo* study showed that aBMD had high discriminatory power for spontaneous vertebral fracture. The cut-off value of aBMD for predicting vertebral fractures without trauma was 0.816 g/cm2, equivalent to -2.62 SD compared to young healthy Japanese women. Low trauma fractures such as a fall from a standing height are due to osteoporosis. The present assessment excluded the traumatic fracture group. Therefore, the threshold value was

This *ex vivo* and *in vivo* study showed that aBMD was a good parameter of vertebral strength and vertebral fracture risk. However, aBMD in isolation can only assess bone density and can not assess bone quality. Therefore, accuracy of assessing vertebral strength by aBMD is

To improve accuracy of assessing vertebral strength and vertebral fracture risk, new method has been developed. CT-based nonlinear FE method can accurately predict vertebral strength, fracture sites and distribution of minimum principal strain *ex vivo* [77]. Based on verification by the cadaver studies, FE method has been applied clinically. A study assessing vertebral fracture risk and medication effects on osteoporosis *in vivo* with CT-based nonlinear FE method showed that analyzed vertebral compressive strength had stronger discriminatory power for vertebral fracture than aBMD and detected alendronate effects at 3

not for diagnosing osteoporosis, but for assessing spontaneous vertebral fracture risk.

measured by DXA to predict spontaneous vertebral fracture.

limited.

months earlier than aBMD [80].

aBMD measurements in isolation might assess vertebral strength well.


*Ex Vivo* and *In Vivo* Assessment of Vertebral Strength

495.

556.

2008;11:163-187.

1996;348:511-514.

Med 1997;157:629-634.

Miner Res 1995;10:341-345.

devices. London, Martin Dunitz. 1999;145-162.

women: A prospective study. BMJ 1990;301:638-641.

and Vertebral Fracture Risk Assessed by Dual Energy X-Ray Absorptiometry 117

[28] Njeh CF, Hans D, Fuerst T, Gluer CC, Genant HK (ed.), Quantitative ultrasound:

[29] Durosier C, Hans D, Krieg MA, Schott AM. Prediction and discrimination of

[30] Krieg MA, Barkmann R, Gonnelli S, *et al.* Quantitative ultrasound in the management of

[31] Porter RW, Miller CG, Grainger D, Palmer SB. Prediction of hip fracture in elderly

[32] Heaney RP, Avioli LV, Chesnut CH 3rd, Lappe J, Recker RR, Brandenburger GH.

[33] Hans D, Dargent-Molina P, Schott AM, *et al*. Ultrasonographic heel measurements to

[34] Bauer DC, Gluer CC, Cauley JA, *et al*. Broadband ultrasound attenuation predicts

[35] Pluijm SM, Graafmans WC, Bouter LM, Lips P. Ultrasound measurements for the

[36] Fujiwara S, Sone T, Yamazaki K, *et al*. Heel bone ultrasound predicts non-spine fracture

[37] Schott AM, Hans D, Duboeuf F, *et al*. Quantitative ultrasound parameters as well as

in elderly women. Results from the EPIDOS study. Bone 2005;37:858-863. [38] Gluer MG, Minne HW, Gluer CC, *et al*. Prospective identification of postmenopausal

[39] Krieg MA, Cornuz J, Ruffieux C, *et al*. Prediction of hip fracture risk by quantitative

[40] Diez-Perez, Gonzalez-Macias, Marin F, *et al*. Prediction of absolute risk of non-spinal

[41] Bauer DC, Ewing S, Cauley J, Ensrud K, Cummings S, Orwoll E. Quantitative

[42] Durosier C, Hans D, Krieg MA, Schott AM. Prediction and discrimination of

in Japanese men and women. Osteoporos int 2005;16:2107-2112.

PIOS study. J Clin Densitom 2005;8:386-395.

Miner Res 2006;21:1457-1463.

Osteoporos Int 2007;18:771-777.

2007;18:629-639.

495.

assessment of osteoporisis and bone status – Non-heel quantitative ultrasound

osteoporotic hip fracture in postmenopausal women. J Clin Densitom 2006;9:475-

osteoporosis: The 2007 ISCD official positions. J Clin Densitom Assess Skelet Health

Ultrasound velocity, through bone predicts incident vertebral deformity. J Bone

predict hip fracture in elderly women: The EPIDOS prospective study. Lancet

fractures strongly and independently of densitometry in older women. A prospective study. Study of Osteoporotic Fractures Research Group. Arch Intern

prediction of osteoporotic fractures in elderly people. Osteoporos Int 1999;9:550-

bone mineral density are better predictors of trochanteric than cervical hip fractures

osteoporotic women at high vertebral fracture risk by radiography, bone densitometry, quantitative ultrasound, and laboratory findings: results from the

ultrasound in more than 7000 Swiss women > or = 70 years of age: Comparison of three technologically different bone ultrasound devices in the SEMOF study. J Bone

fractures using clinical risk factors and heel quantitative ultrasound. Osteoporos Int

ultrasound predicts hip and non-spine fracture in men: The MrOS study.

osteoporotic hip fracture in postmenopausal women. J Clin Densitom 2006;9:475-


[11] Bjarnason K, Hassager C, Svendsen OL, Stang H, Christiansen C. Anteroposterior and

[12] Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 1996;312:1254-1259. [13] Vogel JM. Bone mineral measurement: Skylab experiment M-078. Acta Astronaut

[14] Oyster N, Smith FW. A postmortem correlation of four techniques of assessment of osteoporosis with force of bone compression. Calcif Tissue Int 1988;43:77-82. [15] Uebelhart D, Duboeuf F, Meunier PJ, Delmas PD. Lateral dual-photon absorptiometry: a

[16] Rüegsegger P, Elsasser U, Anliker M, Gnehm H, Kind H, Prader A. Quantification of bone mineralization using computed tomography. Radiology 1976;121:93-97. [17] Isherwood I, Rutherford RA, Pullan BR, Adams PH. Bone-mineral estimation by computer-assisted transverse axial tomography. Lancet 1976;2:712-715. [18] Genant HK, Boyd D, Rosenfeld D, Abols Y, Cann CE. Computed tomography. Non-

[19] Powell MR, Kolb FO, Genant HK, Cann CE, Stebler BG. Comparison of dual photon

[20] McBroom RJ, Hayes WC, Edwards WT, *et al.* Prediction of vertebral body compressive

[21] Cann CE, Genant HK, Kolb FO, Ettinger B. Quantitative computed tomography for

[22] Mosekilde L, Bentzen SM, Ortoft G, Jorgensen J. The predictive value of quantitative

[23] Eriksson SA, Isberg BO, Lindgren JU. Prediction of vertebral strength by dual photon

[24] Kalender WA, Klotz E, Suess C. Vertebral bone mineral analysis: an integrated

[25] Huda W, Morin RL. Patient doses in bone mineral densitometry. Br J Radiol

[26] Hans D, Arlot ME, Schott AM, Roux JP, Kotzki PO, Meunier PJ. Do ultrasound

prediction of vertebral fracture risk. Bone 1985;6:1-7.

approach with CT. Radiology 1987;164:419-423.

London, Martin Dunitz. 1999;109-144.

hip and forearm measurement. Osteoporosis Int 1996;6:37-42.

1975;2:129-139.

Miner Res 1990;5:525-531.

Press 1981;121-149.

A:1206-1214.

Bone 1989;10:465-470.

1989;44:243-250.

1996;69:422-425.

Excerpta Medica 1983;58-60.

lateral spinal DXA for the assessment of vertebral body strength: comparison with

new technique to measure the bone mineral density at the lumbar spine. J Bone

invasive measurements of bone mass and their clinical application. Florida, CRC

absorbtiometry and quantitative computer tomography of the lumbar spine in the same subjects. Clinical disorders of bone and mineral metabolism. Amsterdam,

fracture using quantitative computed tomography. J Bone Joint Surg 1985;67-

computed tomography for vertebral body compressive strength and ash density.

absorptiometry and quantitative computed tomography. Calcif Tissue Int

measurements on the os calcis reflect more the bone microarchitecture than the bone mass? A two-dimensional histomorphometric study. Bone 1995;16:295-300. [27] Njeh CF, Hans D, Fuerst T, Gluer CC, Genant HK (ed.), Quantitative ultrasound:

assessment of osteoporisis and bone status – Calcaneal quantitative ultrasound.


*Ex Vivo* and *In Vivo* Assessment of Vertebral Strength

fracture. Osteoporosis Int 2007;18:1651-1659.

the first decade. J Biomech 1983;16:385-409.

Phys 1997;19:742-747.

Int 1997;7:471-477.

1991;179:669-674.

Res 1998;16:300-308.

2003;33:744-750.

Osteoporos Int 2005;16:642-650.

and Vertebral Fracture Risk Assessed by Dual Energy X-Ray Absorptiometry 119

[61] Han S, Medige J, Faran K, Feng Z, Ziv I. The ability of quantitative ultrasound to predict

[62] Njeh CF, Kuo CW, Langton CM, Atrah HI, Boivin CM. Prediction of human femoral

[63] Durosier C, Hans D, Krieg MA, *et al*. Combining clinical factors and quantitative

[64] Brekelmans WA, Poort HW, Slooff TJ. A new method to analyse the mechanical

[65] Huiskes R, Chao EY. A survey of finite element analysis in orthopedic biomechanics:

[66] Cody DD, Gross GJ, Hou FJ, *et al.* Femoral strength is better predicted by finite element

[67] Keyak JH, Rossi SA, Jones KA, *et al.* Prediction of femoral fracture load using automated

[68] Keyak JH, Rossi SA, Jones KA, *et al.* Prediction of fracture location in the proximal

[69] Keyak JH. Improved prediction of proximal femoral fracture load using nonlinear finite

[70] Bessho M, Ohnishi I, Matsuyama J, *et al.* Prediction of strength and strain of the

[71] Faulkner KG, Cann CE, Hasegawa BH. Effect of bone distribution on vertebral strength:

[72] Silva MJ, Keaveny TM, Hayes WC. Computed tomography-based finite element

[73] Martin H, Werner J, Andresen R, *et al.* Noninvasive assessment of stiffness and failure load of human vertebrae from CT-data. Biomed Tech 1998;43:82-88. [74] Buckley JM, Loo K, Motherway J. Comparison of quantitative computed tomographybased measures in predicting vertebral strength. Bone 2007;40:767-774. [75] Crawford RP, Cann CE, Keaveny TM. Finite element models predict in vitro vertebral

[76] Liebschner MA, Kopperdahl DL, Rosenberg D, *et al.* Finite element modeling of the

[77] Imai K, Ohnishi I, Bessho M, *et al.* Nonlinear finite element model predicts vertebral

[78] Lian KC, Lang TF, Keyak JH, *et al.* Differences in hip quantitative computed

proximal femur by a CT-based finite element method. J Biomech 2007;40:1745-1753.

assessment with patient-specific nonlinear finite element analysis. Radiology

analysis predicts failure loads and fracture patterns for vertebral sections. J Orthop

body compressive strength better than quantitative computed tomography. Bone

tomography (QCT) measurements of bone mineral density and bone strength between glucocorticoid-treated and glucocorticoid-naive postmenopausal women.

femur using finite element models. Med Eng Phys 2001;23:657-664.

behaviour of skeletal parts. Acta Orthop Scand 1972;43:301-317.

models than QCT and DXA. J Biomech 1999;32:1013-1020.

finite element modeling. J Biomech 1998;31:125-133.

element models. Med Eng Phys 2001;23:165-173.

human thoracolumbar spine. Spine 2003;28:559-565.

bone strength and fracture site. Spine 2006;31:1789-1794.

the mechanical properties of trabecular bone under different strain rates. Med Eng

bone strength using ultrasound velocity and BMD. An *in vitro* study. Osteoporosis

ultrasound improves the detection of women both at low and high risk for hip


[43] Frost ML, Blake GM, Fogelman I. Can the WHO criteria for diagnosing osteoporosis be applied to calcaneal quantitative ultrasound? Osteoporosis Int 2000;11:321-330. [44] Faulkner KG, von Stetten E, Miller P. Discordance in patient classification using T-

[45] Damilakis J, Perisinakis K, Gourtsoyiannis N. Imaging ultrasonometry of the calcaneus:

[46] Knapp KM, Blake GM, Spector TD, Fogelman I. Can the WHO definition of

[47] Clowes JA, Peel NF, Eastell R. Device-specific thresholds to diagnose osteoporosis at the

[48] Siffert RS, Kaufman JJ. Ultrasonic bone assessment: The time has come. Bone 2007;40:5-

[49] Roux C, Fournier B, Laugier P, *et al*. Broadband ultrasound attenuation imaging: A new imaging method in osteoporosis. J Bone Miner Res 1996;11:1112-1118. [50] Hans D, Arlot ME, Schott AM, Roux JP, Kotzki PO, Meunier PJ. Do ultrasound

[52] Ammann P, Rizzoli R. Bone strength and its determinants. Osteoporosis Int 2003;14:S13-

[53] Hans D, Wu C, Njeh CF, *et al*. Ultrasound velocity of trabecular cubes reflects mainly

[54] Gluer CC. Quantitative ultrasound - it is time to focus research efforts. Bone 2007;40:9-

[55] Gluer CC, Wu CY, Genant HK. Broadband ultrasound attenuation signals depend on trabecular orientation: An *in vitro* study. Osteoporosis Int 1993;3:185-191. [56] Gluer CC, Wu CY, Jergas M, Goldstein SA, Genant HK. Three quantitative ultrasound

[57] Bouxsein ML, Coan BS, Lee SC. Prediction of the strength of the elderly proximal femur

[58] Bouxsein ML, Radloff SE. Quantitative ultrasound of the calcaneus reflects the

[59] Lochmuller EM, Burklein D, Kuhn V, *et al*. Mechanical strength of the thoracolumbar

computed tomography, and quantitative ultrasound. Bone 2002;31:77-84. [60] Hakulinen MA, Toyras J, Saarakkala S, Hirvonen J, Kroger H, Jurvelin JS. Ability of

by bone mineral density and quantitative ultrasound measurements of the heel and

mechanical properties of calcaneal trabecular bone. J Bone Miner Res 1997;12:839-

spine in the elderly: Prediction from in situ dual-energy X-ray absorptiometry, quantitative computed tomography, upper and lower limb peripheral quantitative

ultrasound backscattering to predict mechanical properties of bovine trabecular

Optimum T-score thresholds for the identification of osteoporotic subjects. Calcif

osteoporosis be applied to multisite axial transmission quantitative ultrasound?

proximal femur: An approach to interpreting peripheral bone measurements in

measurements on the os calcis reflect more the bone microarchitecture than the bone mass?: A two-dimensional histomorphometric study. Bone 1995;16:295-300. [51] Seeman E, Delmas PD. Bone quality - the material and structural basis of bone strength

scores. J Clin Densitom 1999;2:343-350.

Tissue Int 2001;68:219-224.

8.

S18.

13.

846.

tibia. Bone 1999;25:49-54.

Osteoporosis Int 2004;15:367-374.

clinical practice. Osteoporosis Int 2006;17:1293-1302.

and fragility. N Engl J Med 2006;354:2250-2261.

bone. Ultrasound Med Biol 2004;30:919-927.

bone density and elasticity. Calcif Tissue Int 1999;64:18-23.

parameters reflect bone structure. Calcif Tissue Int 1994;55:46-52.


**9** 

*France* 

**Dietary Protein and Bone Health** 

*UMR-914 INRA-AgroParisTech, Nutrition Physiology and Ingestive Behavior, Paris,* 

Dietary proteins represent 10 to 20% of energy consumption. The recommended daily minimum intake of protein and amino acids in adults is 0.8 g per kg of body weight. However, no upper limit has been identified. In industrialized countries, the main sources of protein are milk, eggs and meat. The nutritional value of protein is influenced by several factors, especially the amino acid (AA) composition, protein digestibility, protein digestion kinetics and the ability to transfer AA for protein synthesis. Diets based on either animal or vegetable products supply proteins of differing quality in differing quantities. Plant proteins are often deficient or low in some specific indispensable AAs. Soy protein is reported as a "complete" protein but its overall indispensable AA content is relatively low (~85% lower

Epidemiological data support a positive association between protein intake and bone health. Protein is the precursor of AAs used for bone matrix protein synthesis. Moreover, studies evaluating the relationship between dietary protein and bone turnover support the hypothesis that high protein intake may decrease bone resorption. IGF-1 is a key mediator of bone growth (Geusens & Boonen, 2002) and dietary protein is an important regulator of circulating IGF-1 levels (Bonjour *et al.*, 2001). However, protein is also known to be calciuric, though the origin of the increased calcium excretion is debated. According to the acid ash hypothesis, the protein-induced acid load would have a deleterious effect on bone (Barzel & Massey, 1998). Finally, some proteins, due to their amino acid content, have specific effects on bone metabolism by acting directly on bone cells or through

Even if the effect of nutrition on bone is not as dramatic as that of pharmaceuticals, it can be of great help when considering bone loss. Indeed nutrition is a lower-cost, longer-term, higher-compliance and wider-spread approach than pharmaceutical treatment. The nutritional intervention can be done alone preventively or in combination with a therapy in more severe cases. Some nutritional components such as calcium and vitamin D are recognized to have a positive effect on bone. However, the effect of protein on bone health is

This chapter reviews the literature on the subject, giving an overview of the data available and comparing the proposed mechanisms of action. Attention is drawn to protein quality and to the specific effect of some soy, collagen and milk peptides on bone metabolism.

**1. Introduction** 

indirect pathways.

much more debated.

than milk) (Wilson & Wilson, 2006).

Anne Blais, Emilien Rouy and Daniel Tomé


### **Dietary Protein and Bone Health**

Anne Blais, Emilien Rouy and Daniel Tomé

*UMR-914 INRA-AgroParisTech, Nutrition Physiology and Ingestive Behavior, Paris, France* 

#### **1. Introduction**

120 Dual Energy X-Ray Absorptiometry

[79] Keaveny TM, Donley DW, Hoffmann PF, *et al.* Effects of teriparatide and alendronate on

[80] Imai K, Ohnishi I, Matsumoto T, Yamamoto S, Nakamura K. Assessment of vertebral

[81] Matsumoto T, Ohnishi I, Bessho M, Imai K, Ohashi S, Nakamura K. Prediction of

[82] Bessho M, Ohnishi I, Matsumoto T, *et al.* Prediction of proximal femur strength using a

with osteoporosis. J Bone Miner Res 2007;22:149-157.

Osteoporosis Int 2009;20:801-810.

element model. Spine 2008;33:27-32.

2009;34:1464-1469.

vertebral strength as assessed by finite element modeling of QCT scans in women

fracture risk and therapeutic effects of alendronate in postmenopausal women using a quantitative computed tomography-based nonlinear finite element method.

vertebral strength under loading conditions occurring in activities of daily living using a computed tomography-based nonlinear finite element method. Spine

CT-based nonlinear finite element method: differences in predicted fracture load and site with changing load and boundary conditions. Bone 2009;45:226-231. [83] Imai K, Ohnishi I, Yamamoto S, Nakamura K. *In vivo* assessment of lumbar vertebral

strength in elderly women using computed tomography-based nonlinear finite

Dietary proteins represent 10 to 20% of energy consumption. The recommended daily minimum intake of protein and amino acids in adults is 0.8 g per kg of body weight. However, no upper limit has been identified. In industrialized countries, the main sources of protein are milk, eggs and meat. The nutritional value of protein is influenced by several factors, especially the amino acid (AA) composition, protein digestibility, protein digestion kinetics and the ability to transfer AA for protein synthesis. Diets based on either animal or vegetable products supply proteins of differing quality in differing quantities. Plant proteins are often deficient or low in some specific indispensable AAs. Soy protein is reported as a "complete" protein but its overall indispensable AA content is relatively low (~85% lower than milk) (Wilson & Wilson, 2006).

Epidemiological data support a positive association between protein intake and bone health. Protein is the precursor of AAs used for bone matrix protein synthesis. Moreover, studies evaluating the relationship between dietary protein and bone turnover support the hypothesis that high protein intake may decrease bone resorption. IGF-1 is a key mediator of bone growth (Geusens & Boonen, 2002) and dietary protein is an important regulator of circulating IGF-1 levels (Bonjour *et al.*, 2001). However, protein is also known to be calciuric, though the origin of the increased calcium excretion is debated. According to the acid ash hypothesis, the protein-induced acid load would have a deleterious effect on bone (Barzel & Massey, 1998). Finally, some proteins, due to their amino acid content, have specific effects on bone metabolism by acting directly on bone cells or through indirect pathways.

Even if the effect of nutrition on bone is not as dramatic as that of pharmaceuticals, it can be of great help when considering bone loss. Indeed nutrition is a lower-cost, longer-term, higher-compliance and wider-spread approach than pharmaceutical treatment. The nutritional intervention can be done alone preventively or in combination with a therapy in more severe cases. Some nutritional components such as calcium and vitamin D are recognized to have a positive effect on bone. However, the effect of protein on bone health is much more debated.

This chapter reviews the literature on the subject, giving an overview of the data available and comparing the proposed mechanisms of action. Attention is drawn to protein quality and to the specific effect of some soy, collagen and milk peptides on bone metabolism.

Dietary Protein and Bone Health 123

The effect of protein intake on IGF-1 level could involve Calcium-Sensing Receptor (CaSR) which was also reported to be implicated in the calciuric effect of protein (Conigrave *et al.*, 2008). *In vitro* investigations showed that CaSR is expressed in hepatocytes (Canaff *et al.*, 2001). The activation of CaSR in the liver by AA could explain the increased IGF-1 level due to protein consumption. Moreover, osteoblasts also express CaSR and produce IGF-1 under its regulation. CaSR also upregulates numerous bone forming mechanisms and inhibits bone resorption (Marie, 2010). Aromatic AA supplementation induces an increase in IGF-1 while branched chain AAs have no effect (Dawson-Hughes *et al.*, 2007). This is consistent with the findings of Conigrave *et al*., that aromatic AAs are the most potent activators of

There is no doubt that protein induces an increase of IGF-1 level. By doing so and according to what is known of IGF-1, protein is likely to have an anabolic effect on bone. The proposed mechanism would be that a high protein intake (especially one rich in aromatic AAs) would activate CaSR in hepatocytes and osteoblasts and stimulate IGF-1 production, which would

Bone turnover markers are measured in blood or urine; they give information on bone formation or on bone resorption. Both types of markers are needed in a study in order to

The recent meta-analysis of Darling *et al*. concludes that there is no clear evidence of an effect of protein intake on bone markers (Darling *et al.*, 2009). Some studies reported an increase in bone resorption while bone formation remained stable (Kerstetter *et al.*, 1999; Roughead *et al.*, 2003; Zwart *et al.*, 2005). However, in one of them, the authors report an increase of hydroxyproline (a bone resorption marker) that could be related to the protein source: a collagen-rich meat (Roughead *et al.*, 2003). On the other hand, three studies considering both formation and resorption markers concluded that there was a positive effect of protein on bone. One of them observed a decrease of desoxypyridinoline (a bone resorption marker) with stable formation, indicating a positive decoupling (Hunt *et al.*, 2009). In the second study, both formation and resorption markers increased in a proteinsupplemented group compared to control during 6 months of exercise. According to the authors, bone formation will increase faster than bone resorption over time (Ballard *et al.*, 2005). In the final study, a one year so**y** protein supplementation only increased bone formation markers while resorption markers remained at the same level (Arjmandi *et al.*, 2005). Finally, two intervention studies reported a decrease of resorption with protein supplementation but provided no data on bone formation (Dawson-Hughes *et al.*, 2004; Ince

Urinary calcium excretion is the major pathway for body calcium loss, along with that of feces and sweat. A high urinary calcium level has been linked to increased bone loss

CaSR, while branched chain AAs are the less potent ones (Conigrave *et al.*, 2000).

in turn exert an anabolic effect on bone.

**3. Protein and calcium metabolism** 

**2.3 Bone turnover markers** 

assess bone turnover.

*et al.*, 2004).

**3.1 Urine calcium** 

(Giannini *et al.*, 2003).

#### **2. Action of protein on bone**

The mechanism underlying the debate on the effect of protein on bone is the hormonal anabolic effect. It is based on the increase of bone anabolic hormone IGF-1 with the consumption of protein.

#### **2.1 Bone protein metabolism and turnover**

Type I collagen is the major structural protein distributed throughout the whole body accounting for 25% of total body protein and for 80% of total conjunctive tissue in humans. The different factors involved in bone strength include not only mineral density, crystal characteristics, micro architecture, geometry and morphology (size and shape), but also protein matrix and collagen fiber quality. Cortical bone strength, whose main role is to protect bone integrity, is influenced not only by porosity, presence of micro damage and mineral composition, but also by the orientation of collagen fibers, extent and nature of collagen cross-linking and number and composition of cement lines (Burr, 2002; Currey, 2001; Seeman & Delmas, 2006; Turner, 2006). Collagen is an important component of bone, being the main extra cellular matrix protein for calcification and playing a role in osteoblast differentiation (Takeuchi *et al.*, 1996, 1997).

Markers of bone resorption, which measure the release of peptide resulting of the degradation of mature modified type 1 collagen into the serum and urine, are commonly used to evaluate the relationship between dietary protein and bone turnover. However, to better understand the effect of feeding on bone collagen turnover, a better knowledge of the physiology of bone collagen synthesis is needed (Babraj Smith 2005).

#### **2.2 Hormonal anabolic effect**

Protein has an anabolic effect on bone metabolism by upregulating Insulin-like Growth Factor 1 (IGF-1). IGF-1 is a peptide hormone mainly produced by the liver under the action of the Growth Hormone. A small percentage of IGF-1 is also produced locally by different cell types throughout the body, including osteoblasts (Ohlsson *et al.*, 2009). There is good evidence of the anabolic effect of IGF-1 on bone and muscle, especially when considering age-related disorders (Perrini *et al.*, 2010). Studies measuring IGF-1 levels under different protein intakes consistently report that a high protein intake increases IGF-1. This fact has been shown over a six-month period in young exercising subjects (Ballard *et al.*, 2005) and evidence also exists in older populations of both sexes (Arjmandi *et al.*, 2005; Dawson-Hughes *et al.*, 2004; Hunt *et al.*, 2009).

There is some evidence that this hormonal response is based on the quality of the protein ingested. Dawson-Hughes and colleagues showed that a fivefold increase of aromatic AA intake (phenylalanine and histidine) for 24 days induced an increase of IGF-1 level. On the other hand, a fivefold increase of branched chain AA (leucine and isoleuine) induced no effect on IGF-1, indicating that the protein intake effect on IGF-1 is probably AA-dependent (Dawson-Hughes *et al.*, 2007). Moreover, two studies comparing milk and soy protein reported that soy induces a greater increase of IGF-1 (Arjmandi *et al.*, 2003; Khalil *et al.*, 2002). However, it is important to note that the soy protein contained isoflavone in both studies, thus the effect on IGF-1 could be attributed to this molecule rather than to the AA content.

The mechanism underlying the debate on the effect of protein on bone is the hormonal anabolic effect. It is based on the increase of bone anabolic hormone IGF-1 with the

Type I collagen is the major structural protein distributed throughout the whole body accounting for 25% of total body protein and for 80% of total conjunctive tissue in humans. The different factors involved in bone strength include not only mineral density, crystal characteristics, micro architecture, geometry and morphology (size and shape), but also protein matrix and collagen fiber quality. Cortical bone strength, whose main role is to protect bone integrity, is influenced not only by porosity, presence of micro damage and mineral composition, but also by the orientation of collagen fibers, extent and nature of collagen cross-linking and number and composition of cement lines (Burr, 2002; Currey, 2001; Seeman & Delmas, 2006; Turner, 2006). Collagen is an important component of bone, being the main extra cellular matrix protein for calcification and playing a role in osteoblast

Markers of bone resorption, which measure the release of peptide resulting of the degradation of mature modified type 1 collagen into the serum and urine, are commonly used to evaluate the relationship between dietary protein and bone turnover. However, to better understand the effect of feeding on bone collagen turnover, a better knowledge of the

Protein has an anabolic effect on bone metabolism by upregulating Insulin-like Growth Factor 1 (IGF-1). IGF-1 is a peptide hormone mainly produced by the liver under the action of the Growth Hormone. A small percentage of IGF-1 is also produced locally by different cell types throughout the body, including osteoblasts (Ohlsson *et al.*, 2009). There is good evidence of the anabolic effect of IGF-1 on bone and muscle, especially when considering age-related disorders (Perrini *et al.*, 2010). Studies measuring IGF-1 levels under different protein intakes consistently report that a high protein intake increases IGF-1. This fact has been shown over a six-month period in young exercising subjects (Ballard *et al.*, 2005) and evidence also exists in older populations of both sexes (Arjmandi *et al.*, 2005; Dawson-

There is some evidence that this hormonal response is based on the quality of the protein ingested. Dawson-Hughes and colleagues showed that a fivefold increase of aromatic AA intake (phenylalanine and histidine) for 24 days induced an increase of IGF-1 level. On the other hand, a fivefold increase of branched chain AA (leucine and isoleuine) induced no effect on IGF-1, indicating that the protein intake effect on IGF-1 is probably AA-dependent (Dawson-Hughes *et al.*, 2007). Moreover, two studies comparing milk and soy protein reported that soy induces a greater increase of IGF-1 (Arjmandi *et al.*, 2003; Khalil *et al.*, 2002). However, it is important to note that the soy protein contained isoflavone in both studies, thus the effect on IGF-1 could be attributed to this molecule rather than to the AA

physiology of bone collagen synthesis is needed (Babraj Smith 2005).

**2. Action of protein on bone** 

**2.1 Bone protein metabolism and turnover** 

differentiation (Takeuchi *et al.*, 1996, 1997).

**2.2 Hormonal anabolic effect** 

Hughes *et al.*, 2004; Hunt *et al.*, 2009).

content.

consumption of protein.

The effect of protein intake on IGF-1 level could involve Calcium-Sensing Receptor (CaSR) which was also reported to be implicated in the calciuric effect of protein (Conigrave *et al.*, 2008). *In vitro* investigations showed that CaSR is expressed in hepatocytes (Canaff *et al.*, 2001). The activation of CaSR in the liver by AA could explain the increased IGF-1 level due to protein consumption. Moreover, osteoblasts also express CaSR and produce IGF-1 under its regulation. CaSR also upregulates numerous bone forming mechanisms and inhibits bone resorption (Marie, 2010). Aromatic AA supplementation induces an increase in IGF-1 while branched chain AAs have no effect (Dawson-Hughes *et al.*, 2007). This is consistent with the findings of Conigrave *et al*., that aromatic AAs are the most potent activators of CaSR, while branched chain AAs are the less potent ones (Conigrave *et al.*, 2000).

There is no doubt that protein induces an increase of IGF-1 level. By doing so and according to what is known of IGF-1, protein is likely to have an anabolic effect on bone. The proposed mechanism would be that a high protein intake (especially one rich in aromatic AAs) would activate CaSR in hepatocytes and osteoblasts and stimulate IGF-1 production, which would in turn exert an anabolic effect on bone.

#### **2.3 Bone turnover markers**

Bone turnover markers are measured in blood or urine; they give information on bone formation or on bone resorption. Both types of markers are needed in a study in order to assess bone turnover.

The recent meta-analysis of Darling *et al*. concludes that there is no clear evidence of an effect of protein intake on bone markers (Darling *et al.*, 2009). Some studies reported an increase in bone resorption while bone formation remained stable (Kerstetter *et al.*, 1999; Roughead *et al.*, 2003; Zwart *et al.*, 2005). However, in one of them, the authors report an increase of hydroxyproline (a bone resorption marker) that could be related to the protein source: a collagen-rich meat (Roughead *et al.*, 2003). On the other hand, three studies considering both formation and resorption markers concluded that there was a positive effect of protein on bone. One of them observed a decrease of desoxypyridinoline (a bone resorption marker) with stable formation, indicating a positive decoupling (Hunt *et al.*, 2009). In the second study, both formation and resorption markers increased in a proteinsupplemented group compared to control during 6 months of exercise. According to the authors, bone formation will increase faster than bone resorption over time (Ballard *et al.*, 2005). In the final study, a one year so**y** protein supplementation only increased bone formation markers while resorption markers remained at the same level (Arjmandi *et al.*, 2005). Finally, two intervention studies reported a decrease of resorption with protein supplementation but provided no data on bone formation (Dawson-Hughes *et al.*, 2004; Ince *et al.*, 2004).

#### **3. Protein and calcium metabolism**

#### **3.1 Urine calcium**

Urinary calcium excretion is the major pathway for body calcium loss, along with that of feces and sweat. A high urinary calcium level has been linked to increased bone loss (Giannini *et al.*, 2003).

Dietary Protein and Bone Health 125

The effect of protein on calcium absorption could also be conditioned by the calcium intake itself. Two studies tested the effect of protein intake on intestinal calcium absorption at different levels of calcium intake. One observed a positive effect of protein for 675 mg Ca/d and no effect at 1510 mg Ca/d (Hunt *et al.*, 2009) whereas an other study observed a

Moreover, Kerstetter et al. repeatedly showed that, in the context of low protein intake, there is both an increase of PTH and a decrease in calcium absorption (Kerstetter *et al.*, 1997, 1998, 2006). Those two findings seem incompatible because PTH is known to increase calcium absorption (Heaney, 2007). However, Conigrave *et al*., as a part of his theory on amino acid-sensing CaSR, proposes another mechanism not involving PTH. CaSR is also present in the gastro-intestinal tract, where it promotes gastric acid secretion directly and indirectly through gastrin release (Conigrave & Brown, 2006). High-protein diets induce the acidification of the intestinal content which helps calcium salts to dissolve into Ca2+, thereby

According to the acid-ash theory, some foods such as protein would induce a metabolic acidosis whereas other foods such as vegetables and fruits would counteract this process. The acidosis resulting from a high protein diet would be deleterious to bone and induce

Dietary proteins, and more specifically their sulphur AA content, are part of the acidforming nutrients. Consequently, high protein consumption would lead to metabolic acidosis (Frassetto *et al.*, 1998; Kerstetter *et al.*, 2006). pH is closely regulated in the body and any increase of the acid load is buffered through different pathways. The main one is through lung excretion of CO2, followed by kidney acid excretion. Bone has also been proposed to take part by releasing Ca2+ (Lemann *et al.*, 2003). This last pathway is debated because the lungs' and the kidneys' buffering capacity are considered to be sufficient to compensate for the protein-induced acidosis (Bonjour, 2005). The idea that an acidic diet is deleterious for bone is supported by observational studies linking acid production to a reduced BMD (New *et al.*, 2004; Rahbar *et al.*, 2009; Wynn *et al.*, 2008). Intervention studies on this topic usually correct the acid diet of the subjects by supplementing them with alkalinizing molecules. The results consistently show an improvement of bone health with the supplementation. This improvement occurs at the level of urinary calcium excretion (Frassetto *et al.*, 2005; Lemann *et al.*, 1991), bone turnover markers (Dawson-Hughes *et al.*, 2009; He *et al.*, 2010; Maurer *et al.*, 2003) or Bone Mineral Density (BMD) (Domrongkitchaiporn *et al.*, 2002). Despite all this evidence of the beneficial effect of alkalinizing the diet, it should be noted that according to a recent meta-analysis, no causal link could be established between dietary acid load and bone disease (Fenton *et al.*, 2011). It is also important to consider that the findings supporting the acid-ash hypothesis are based

negative effect at 871 mg/d and no effect at 1346 mg/d.

**3.3 The acid-ash theory** 

bone loss.

facilitating its absorption even if PTH level is low (Conigrave *et al.*, 2008).

on an alkalinisation of the diet and are not directly related to protein intake.

A decrease of pH is probably one of the main activators of bone-resorbing osteoclasts and an inhibitor of bone matrix deposition by osteoblasts, which explains the acid-induced bone reabsorption (Arnett, 2008). Extrapolation of these results to the clinical level is risky as it compares the *in vitro* pH variation to a high protein-induced acidosis. Indeed, the pH

Numerous studies have reported a positive linear relationship between dietary protein and urinary calcium (Kerstetter & Allen, 1990; Kerstetter *et al.*, 2003; Whiting *et al.*, 1997). The rise in urinary calcium with protein intake has also been observed in more recent studies (Ceglia *et al.*, 2009; Hunt *et al.*, 2009; Jajoo *et al.*, 2006). In one study, no effect of protein on calciuria was observed (Roughead *et al.*, 2003). The mechanism underlying the effect of protein on calcium excretion is debated. First, a high protein diet induces an increase of glomerular filtration rate which will exacerbate any increased calcium excretion (Kerstetter *et al.*, 1998). According to the acid ash hypothesis, the calciuric effect of a high protein intake would be caused by the protein-induced acid load. A meta-analysis by Fenton *et al*. of over 25 studies concluded that changes in acid excretion modulate calcium excretion (Fenton *et al.*, 2008). Other studies showed that urinary calcium could be reduced by decreasing the acid load (L. Frassetto *et al.*, 2005; Lemann *et al.*, 1991). The systemic acidosis generated by the protein load inhibits TRPV5 calcium channel expression in the kidney. Because TRPV5 participates in renal calcium reabsorption, its inhibition increases urinary calcium (Nijenhuis *et al.*, 2006).

Another mechanism for the protein-induced calcium excretion was proposed by Conigrave and colleagues, involving the extracellular CaSR (Conigrave *et al.*, 2008). CaSR found on cells of the parathyroid glands and in the renal tubules are sensitive not only to calcium, but also to some AAs (Conigrave *et al.*, 2000). Through this double sensitivity, a link is formed between protein and calcium metabolism. According to these findings, AAs would activate the CaSRs, reducing parathormone (PTH) production and decreasing renal calcium reabsorption, leading eventually to an increased calcium excretion (Conigrave *et al.*, 2008). Indeed, PTH level has been shown to be influenced by protein intake (Kerstetter *et al.*, 1997, 1998, 2006).

Urinary calcium loss due to a high protein diet was attributed to enhanced bone resorption (Kerstetter & Allen, 1990); however, this view has been discussed as the link between protein and bone resorption is unclear (Bonjour, 2005; M. Thorpe & Evans, 2011). Urinary calcium excretion is mainly related to bone as the main calcium storage compartment. However, an increase of the excreted calcium could also be related to an increase in calcium absorption through the diet. Hence the effect of protein on intestinal calcium absorption should also be considered.

#### **3.2 Calcium absorption**

While many studies claim that high-protein diets cause calciuria, there is no clear relationship between protein intake and calcium absorption. Some studies report a positive effect of protein on calcium absorption (Cao *et al.*, 2011; Hunt *et al.*, 2009; Kerstetter *et al.*, 1998, 2005). One study of 13 women observed an increase in calcium absorption that explained 93% of the calcium excretion (Kerstetter *et al.*, 2005). Another study reported an increased calciuria and a non-significant increase in calcium absorption with protein intake. But the net difference between calcium absorption and excretion with high or low protein intake remained stable (Cao *et al.*, 2011). However, other studies observed a decrease of calcium absorption with protein (Dawson-Hughes & Harris, 2002; Heaney, 2000), or no effect at all (Roughead *et al.*, 2003). It should be pointed out that all studies showing a positive effect were short term studies (2 to 7 weeks) whereas the two studies with a negative effect were long-term studies (years). It is thus possible that protein increases calcium absorption at first and that this effect is reversed in the long run.

Numerous studies have reported a positive linear relationship between dietary protein and urinary calcium (Kerstetter & Allen, 1990; Kerstetter *et al.*, 2003; Whiting *et al.*, 1997). The rise in urinary calcium with protein intake has also been observed in more recent studies (Ceglia *et al.*, 2009; Hunt *et al.*, 2009; Jajoo *et al.*, 2006). In one study, no effect of protein on calciuria was observed (Roughead *et al.*, 2003). The mechanism underlying the effect of protein on calcium excretion is debated. First, a high protein diet induces an increase of glomerular filtration rate which will exacerbate any increased calcium excretion (Kerstetter *et al.*, 1998). According to the acid ash hypothesis, the calciuric effect of a high protein intake would be caused by the protein-induced acid load. A meta-analysis by Fenton *et al*. of over 25 studies concluded that changes in acid excretion modulate calcium excretion (Fenton *et al.*, 2008). Other studies showed that urinary calcium could be reduced by decreasing the acid load (L. Frassetto *et al.*, 2005; Lemann *et al.*, 1991). The systemic acidosis generated by the protein load inhibits TRPV5 calcium channel expression in the kidney. Because TRPV5 participates in renal calcium reabsorption, its inhibition increases urinary calcium (Nijenhuis *et al.*, 2006). Another mechanism for the protein-induced calcium excretion was proposed by Conigrave and colleagues, involving the extracellular CaSR (Conigrave *et al.*, 2008). CaSR found on cells of the parathyroid glands and in the renal tubules are sensitive not only to calcium, but also to some AAs (Conigrave *et al.*, 2000). Through this double sensitivity, a link is formed between protein and calcium metabolism. According to these findings, AAs would activate the CaSRs, reducing parathormone (PTH) production and decreasing renal calcium reabsorption, leading eventually to an increased calcium excretion (Conigrave *et al.*, 2008). Indeed, PTH level has been shown to be influenced by protein intake (Kerstetter *et al.*, 1997,

Urinary calcium loss due to a high protein diet was attributed to enhanced bone resorption (Kerstetter & Allen, 1990); however, this view has been discussed as the link between protein and bone resorption is unclear (Bonjour, 2005; M. Thorpe & Evans, 2011). Urinary calcium excretion is mainly related to bone as the main calcium storage compartment. However, an increase of the excreted calcium could also be related to an increase in calcium absorption through the diet. Hence the effect of protein on intestinal calcium absorption

While many studies claim that high-protein diets cause calciuria, there is no clear relationship between protein intake and calcium absorption. Some studies report a positive effect of protein on calcium absorption (Cao *et al.*, 2011; Hunt *et al.*, 2009; Kerstetter *et al.*, 1998, 2005). One study of 13 women observed an increase in calcium absorption that explained 93% of the calcium excretion (Kerstetter *et al.*, 2005). Another study reported an increased calciuria and a non-significant increase in calcium absorption with protein intake. But the net difference between calcium absorption and excretion with high or low protein intake remained stable (Cao *et al.*, 2011). However, other studies observed a decrease of calcium absorption with protein (Dawson-Hughes & Harris, 2002; Heaney, 2000), or no effect at all (Roughead *et al.*, 2003). It should be pointed out that all studies showing a positive effect were short term studies (2 to 7 weeks) whereas the two studies with a negative effect were long-term studies (years). It is thus possible that protein increases

calcium absorption at first and that this effect is reversed in the long run.

1998, 2006).

should also be considered.

**3.2 Calcium absorption** 

The effect of protein on calcium absorption could also be conditioned by the calcium intake itself. Two studies tested the effect of protein intake on intestinal calcium absorption at different levels of calcium intake. One observed a positive effect of protein for 675 mg Ca/d and no effect at 1510 mg Ca/d (Hunt *et al.*, 2009) whereas an other study observed a negative effect at 871 mg/d and no effect at 1346 mg/d.

Moreover, Kerstetter et al. repeatedly showed that, in the context of low protein intake, there is both an increase of PTH and a decrease in calcium absorption (Kerstetter *et al.*, 1997, 1998, 2006). Those two findings seem incompatible because PTH is known to increase calcium absorption (Heaney, 2007). However, Conigrave *et al*., as a part of his theory on amino acid-sensing CaSR, proposes another mechanism not involving PTH. CaSR is also present in the gastro-intestinal tract, where it promotes gastric acid secretion directly and indirectly through gastrin release (Conigrave & Brown, 2006). High-protein diets induce the acidification of the intestinal content which helps calcium salts to dissolve into Ca2+, thereby facilitating its absorption even if PTH level is low (Conigrave *et al.*, 2008).

#### **3.3 The acid-ash theory**

According to the acid-ash theory, some foods such as protein would induce a metabolic acidosis whereas other foods such as vegetables and fruits would counteract this process. The acidosis resulting from a high protein diet would be deleterious to bone and induce bone loss.

Dietary proteins, and more specifically their sulphur AA content, are part of the acidforming nutrients. Consequently, high protein consumption would lead to metabolic acidosis (Frassetto *et al.*, 1998; Kerstetter *et al.*, 2006). pH is closely regulated in the body and any increase of the acid load is buffered through different pathways. The main one is through lung excretion of CO2, followed by kidney acid excretion. Bone has also been proposed to take part by releasing Ca2+ (Lemann *et al.*, 2003). This last pathway is debated because the lungs' and the kidneys' buffering capacity are considered to be sufficient to compensate for the protein-induced acidosis (Bonjour, 2005). The idea that an acidic diet is deleterious for bone is supported by observational studies linking acid production to a reduced BMD (New *et al.*, 2004; Rahbar *et al.*, 2009; Wynn *et al.*, 2008). Intervention studies on this topic usually correct the acid diet of the subjects by supplementing them with alkalinizing molecules. The results consistently show an improvement of bone health with the supplementation. This improvement occurs at the level of urinary calcium excretion (Frassetto *et al.*, 2005; Lemann *et al.*, 1991), bone turnover markers (Dawson-Hughes *et al.*, 2009; He *et al.*, 2010; Maurer *et al.*, 2003) or Bone Mineral Density (BMD) (Domrongkitchaiporn *et al.*, 2002). Despite all this evidence of the beneficial effect of alkalinizing the diet, it should be noted that according to a recent meta-analysis, no causal link could be established between dietary acid load and bone disease (Fenton *et al.*, 2011). It is also important to consider that the findings supporting the acid-ash hypothesis are based on an alkalinisation of the diet and are not directly related to protein intake.

A decrease of pH is probably one of the main activators of bone-resorbing osteoclasts and an inhibitor of bone matrix deposition by osteoblasts, which explains the acid-induced bone reabsorption (Arnett, 2008). Extrapolation of these results to the clinical level is risky as it compares the *in vitro* pH variation to a high protein-induced acidosis. Indeed, the pH

Dietary Protein and Bone Health 127

men. The study lasted 28 days and the authors reported a decrease in BMC in the supplemented subjects whereas BMC in control subjects remained stable (Zwart *et al.*, 2005). It should be noted that energy intake was different between the two groups due to an energy-free placebo. Another study supplemented hospitalized elderly men and women for 38 days with 20.4g of protein. Along with other positive effects (lower complication rate, shorter hospital stay), the authors reported a decrease in BMD loss with the protein supplement (Tkatch *et al.*, 1992). Finally, a year-long study supplementing 62 postmenopausal women with 25g of soy protein observed no effect on total hip BMD or

Fracture risk is the ultimate clinical outcome when considering bone. A longitudinal study of 32 050 postmenopausal women observed a reduced relative risk of hip fracture when protein intake was increased (Munger *et al.*, 1999). Another one observed that a reduction in wrist fracture is positively associated with the consumption frequency of high protein food in 1865 perimenopausal women (D. L. Thorpe *et al.*, 2008a). Finally, in a case-control study on 1167 cases of hip fractures and 1334 controls of both sexes aged 50-89, the odds ratio for

According to these three studies, protein would have a beneficial effect on bone, but some other studies found conflicting results (Dargent-Molina *et al.*, 2008; Feskanich *et al.*, 1996; Meyer *et al.*, 1997). However, it should be noted that the negative relationship between protein and fracture risk was significant only in the lowest quartiles of calcium intake (<400 mg/d) and that no association was observed for higher calcium intake (Dargent-Molina *et al.*, 2008; Meyer *et al.*, 1997). Pooling four studies in a meta-analysis, Darling found no significant effect of protein on the risk ratio of fracture (Darling *et al.*, 2009). To our knowledge, it seems that as long as calcium intake is sufficient there is a protective effect of

The question of the effect of protein on fracture healing has also been addressed in some randomised controlled trials. All four trials were conducted by the same group and reported that a protein supplement improved the patients' condition after a low-trauma femoral neck fracture (Delmi *et al.*, 1990; Tkatch *et al.*, 1992) or hip fracture (Chevalley *et al.*, 2008; Schurch *et al.*, 1998). All the clinical trials had very similar design: patients were about 80 years old and the protein supplement given was 20g. According to the two studies on femoral neck fracture, patients taking the protein supplementation had better clinical outcomes and reduced rate of complication and mortality during the hospital stay and 6 months after (Delmi *et al.*, 1990; Tkatch *et al.*, 1992). After hip fracture, the protein supplement increased IGF-1, attenuating bone loss at 6 months and shortening hospital stay (Schurch *et al.*, 1998). The IGF-1 increase is independent of the type of protein given (casein, whey protein or whey protein and amino acids) (Chevalley *et al.*, 2010). The beneficial effect of protein is probably linked to both the

Each protein bares some information in the form of its chain of amino acids. This information can relate to specific effects of some proteins. When protein goes through the

increased IGF-1 level and the correction of the patient nutritional state.

**5. Effect of specific protein source on bone** 

hip fracture decreased with the increase of protein intake (Wengreen *et al.*, 2004).

BMC (Arjmandi *et al.*, 2005).

protein on bone that lowers the fracture risk.

**4.2 Fracture risk** 

variation induced by protein intake is likely to be too small to induce any effect in the bone micro-environment (Bonjour, 2005).

Although they lead to different conclusions, the acid-ash theory and the hormonal anabolic theory are not mutually exclusive. A dual-pathway model for the effect of protein on bone has been proposed (M. Thorpe & Evans, 2011). Although it has been recently criticized in a meta-analysis (Fenton *et al.*, 2011), it seems that there is a beneficial effect of diet alkalinisation. This can be achieved by lowering acidic or by increasing alkaline food consumption. When considering protein, the second method is more beneficial. Indeed, even if protein is an acid nutrient, it is also an activator of bone anabolic hormones and its consumption should not be lowered. This is especially true in populations already consuming a low protein diet such as the elderly. The acid load resulting from the protein consumption should be compensated by an increased consumption of alkalinizing fruits and vegetables.

Some diets designed to promote weight loss rely on high protein consumption for a quick effect on weight. It is hard to estimate what is the effect of such diets on bone. Indeed, hyperproteic diets imply also potentially confounding effect such as spontaneous caloric restriction and a possible lack of micronutrients. Moreover, the wide variety of hyperproteic diets and the potential lack of compliance of the subjects further increase the difficulty of designing an adequate study on the subject.

#### **4. Protein and bone parameters**

#### **4.1 Bone Mineral Density**

BMD is not the only determinant of skeletal fragility; the spatial distribution of the bone mass (as cortical and trabecular bone) and the intrinsic material properties of bone are also major components (Bouxsein & Seeman, 2009). Measuring BMD or Bone Mineral Content (BMC) is the easiest way of directly assessing bone strength in humans.

Most observation studies report a positive association between protein and bone density (Hannan *et al.*, 2000; Promislow *et al.*, 2002) or between protein and BMD change (Dawson-Hughes & Harris, 2002; Vatanparast *et al.*, 2007). Specific studies showing a positive correlation between protein and bone density cover a broad population: premenopausal women (Teegarden *et al.*, 1998), postmenopausal women (Devine *et al.*, 2005; Ilich *et al.*, 2003; Rapuri *et al.*, 2003; M. Thorpe *et al.*, 2008b), men (Whiting *et al.*, 2002) and children (Alexy *et al.*, 2005; Chevalley *et al.*, 2008).

On the other hand, two studies on premenopausal women concluded that protein intake had no relation with BMD (Beasley *et al.*, 2010; Mazess & Barden, 1991). One study reported a negative association with BMC (Metz *et al.*, 1993). Finally, a case-control study compared 134 osteoporotic women with 137 controls and identified total protein intake as a risk factor for osteoporosis. A meta-analysis by Darling and colleagues conclude that there is no evidence of a negative effect of protein intake on bone when looking at observation studies; in fact there is probably a small positive effect of protein on bone (Darling *et al.*, 2009).

There are very few interventional studies comparing high vs. low protein intake. One of them focused on the effect of an AA and carbohydrate supplement during bed rest on 13

variation induced by protein intake is likely to be too small to induce any effect in the bone

Although they lead to different conclusions, the acid-ash theory and the hormonal anabolic theory are not mutually exclusive. A dual-pathway model for the effect of protein on bone has been proposed (M. Thorpe & Evans, 2011). Although it has been recently criticized in a meta-analysis (Fenton *et al.*, 2011), it seems that there is a beneficial effect of diet alkalinisation. This can be achieved by lowering acidic or by increasing alkaline food consumption. When considering protein, the second method is more beneficial. Indeed, even if protein is an acid nutrient, it is also an activator of bone anabolic hormones and its consumption should not be lowered. This is especially true in populations already consuming a low protein diet such as the elderly. The acid load resulting from the protein consumption should be compensated by an increased consumption of alkalinizing fruits and

Some diets designed to promote weight loss rely on high protein consumption for a quick effect on weight. It is hard to estimate what is the effect of such diets on bone. Indeed, hyperproteic diets imply also potentially confounding effect such as spontaneous caloric restriction and a possible lack of micronutrients. Moreover, the wide variety of hyperproteic diets and the potential lack of compliance of the subjects further increase the difficulty of

BMD is not the only determinant of skeletal fragility; the spatial distribution of the bone mass (as cortical and trabecular bone) and the intrinsic material properties of bone are also major components (Bouxsein & Seeman, 2009). Measuring BMD or Bone Mineral Content

Most observation studies report a positive association between protein and bone density (Hannan *et al.*, 2000; Promislow *et al.*, 2002) or between protein and BMD change (Dawson-Hughes & Harris, 2002; Vatanparast *et al.*, 2007). Specific studies showing a positive correlation between protein and bone density cover a broad population: premenopausal women (Teegarden *et al.*, 1998), postmenopausal women (Devine *et al.*, 2005; Ilich *et al.*, 2003; Rapuri *et al.*, 2003; M. Thorpe *et al.*, 2008b), men (Whiting *et al.*, 2002) and children (Alexy *et* 

On the other hand, two studies on premenopausal women concluded that protein intake had no relation with BMD (Beasley *et al.*, 2010; Mazess & Barden, 1991). One study reported a negative association with BMC (Metz *et al.*, 1993). Finally, a case-control study compared 134 osteoporotic women with 137 controls and identified total protein intake as a risk factor for osteoporosis. A meta-analysis by Darling and colleagues conclude that there is no evidence of a negative effect of protein intake on bone when looking at observation studies; in fact there is probably a small positive effect of protein on bone (Darling *et al.*, 2009).

There are very few interventional studies comparing high vs. low protein intake. One of them focused on the effect of an AA and carbohydrate supplement during bed rest on 13

(BMC) is the easiest way of directly assessing bone strength in humans.

micro-environment (Bonjour, 2005).

designing an adequate study on the subject.

**4. Protein and bone parameters** 

**4.1 Bone Mineral Density** 

*al.*, 2005; Chevalley *et al.*, 2008).

vegetables.

men. The study lasted 28 days and the authors reported a decrease in BMC in the supplemented subjects whereas BMC in control subjects remained stable (Zwart *et al.*, 2005). It should be noted that energy intake was different between the two groups due to an energy-free placebo. Another study supplemented hospitalized elderly men and women for 38 days with 20.4g of protein. Along with other positive effects (lower complication rate, shorter hospital stay), the authors reported a decrease in BMD loss with the protein supplement (Tkatch *et al.*, 1992). Finally, a year-long study supplementing 62 postmenopausal women with 25g of soy protein observed no effect on total hip BMD or BMC (Arjmandi *et al.*, 2005).

#### **4.2 Fracture risk**

Fracture risk is the ultimate clinical outcome when considering bone. A longitudinal study of 32 050 postmenopausal women observed a reduced relative risk of hip fracture when protein intake was increased (Munger *et al.*, 1999). Another one observed that a reduction in wrist fracture is positively associated with the consumption frequency of high protein food in 1865 perimenopausal women (D. L. Thorpe *et al.*, 2008a). Finally, in a case-control study on 1167 cases of hip fractures and 1334 controls of both sexes aged 50-89, the odds ratio for hip fracture decreased with the increase of protein intake (Wengreen *et al.*, 2004).

According to these three studies, protein would have a beneficial effect on bone, but some other studies found conflicting results (Dargent-Molina *et al.*, 2008; Feskanich *et al.*, 1996; Meyer *et al.*, 1997). However, it should be noted that the negative relationship between protein and fracture risk was significant only in the lowest quartiles of calcium intake (<400 mg/d) and that no association was observed for higher calcium intake (Dargent-Molina *et al.*, 2008; Meyer *et al.*, 1997). Pooling four studies in a meta-analysis, Darling found no significant effect of protein on the risk ratio of fracture (Darling *et al.*, 2009). To our knowledge, it seems that as long as calcium intake is sufficient there is a protective effect of protein on bone that lowers the fracture risk.

The question of the effect of protein on fracture healing has also been addressed in some randomised controlled trials. All four trials were conducted by the same group and reported that a protein supplement improved the patients' condition after a low-trauma femoral neck fracture (Delmi *et al.*, 1990; Tkatch *et al.*, 1992) or hip fracture (Chevalley *et al.*, 2008; Schurch *et al.*, 1998). All the clinical trials had very similar design: patients were about 80 years old and the protein supplement given was 20g. According to the two studies on femoral neck fracture, patients taking the protein supplementation had better clinical outcomes and reduced rate of complication and mortality during the hospital stay and 6 months after (Delmi *et al.*, 1990; Tkatch *et al.*, 1992). After hip fracture, the protein supplement increased IGF-1, attenuating bone loss at 6 months and shortening hospital stay (Schurch *et al.*, 1998). The IGF-1 increase is independent of the type of protein given (casein, whey protein or whey protein and amino acids) (Chevalley *et al.*, 2010). The beneficial effect of protein is probably linked to both the increased IGF-1 level and the correction of the patient nutritional state.

#### **5. Effect of specific protein source on bone**

Each protein bares some information in the form of its chain of amino acids. This information can relate to specific effects of some proteins. When protein goes through the

Dietary Protein and Bone Health 129

results because of the broad variety of foods in these groups. In future research, it would be more accurate to consider an estimation of the sulphate content of each protein-containing

Milk contains many bioactive factors, including growth hormones, enzymes, antimicrobials, anti-inflammatory agents, transporters and peptide or nonpeptide hormones. Milk, because it contains bioactive molecules, extends beyond applications in infant nutrition and was

Clinical trials showed that MBP supplementation increased BMD and decreased bone resorption biomarkers in healthy women (Aoe *et al.*, 2001; Uenishi *et al.*, 2007; Yamamura *et al.*, 2002), menopausal women (Aoe *et al.*, 2005), and healthy older women (Aoyagi *et al.*, 2010). In particular, MBP suppresses osteoclast-mediated bone resorption and leads to reduced osteoclast number in animal studies (Morita *et al.*, 2008). Morita *et al*. reported that the protein fraction responsible for the observed activities of MBP is the bovine angiogenin

However, not all studies have shown a beneficial effect of MBP. In one study, 84 healthy young women were divided into three groups receiving placebo, whole milk, or milk containing 40 mg MBP for 8 months. Compared with the baseline values, total BMD significantly increased in all groups. There was a significant decrease of bone resorption marker N-teleopeptides of type-I collagen (NTx) while bone formation remained stable in

HPLC analysis of the major proteins in the MBP fraction identified the presence of the

Lactoferrin (LF) is an 80 kDa iron-binding glycoprotein of the transferrin family. This molecule has been demonstrated to inhibit *in vitro* osteoclast-mediated bone resorption (Lorget *et al.*, 2002). LF was also demonstrated to have *in vitro* anabolic, differentiating and anti-apoptotic effects on osteoblasts, and to inhibit osteoclastogenesis. Moreover *in vivo* local injection of LF above the hemicalvaria increases bone formation and bone area in adult mice (Cornish *et al.*, 2004). LF has a role in host non-specific defense (Gahr *et al.*, 1991; Legrand *et al.*, 2004). In addition to its direct antimicrobial effects, LF is believed to modulate the inflammatory process mainly by preventing the release of inflammatory cytokines which induce recruitment and

Investigations of our group and others established that LF at physiological concentrations can stimulate proliferation of primary osteoblasts and osteoblastic-cell lines and increase osteoblast differentiation (Blais *et al.*, 2009; Cornish, 2004; Takayama & Mizumachi, 2008). Studies using 3-week culture of primary rat osteoblasts show that LF dose-dependently increases the number of nodules and the area of mineralized bone formed (Cornish, 2004). Our *in vitro* experiments demonstrated that LF could directly act on bone cells. Bovine LF

food rather than focusing only on a crude animal-vegetable distinction.

considered as a possible source of factors with anabolic effects on bone.

which acts as a bone resorption-inhibitory protein (Morita *et al.*, 2008).

glycoprotein lactoferrin in most of these fractions (Naot *et al.*, 2005).

activation of immune cells at inflammatory sites (Legrand *et al.*, 2005).

**5.2 Milk proteins** 

**5.2.1 Milk Basic Proteins (MBP)** 

both milk groups. (Zou *et al.*, 2009).

**5.2.2 Lactoferrin** 

intestinal barrier, 50% are completely degraded as simple amino acids, 40% are partly degraded as peptides and 10% remain intact (Mallegol *et al.*, 2005). This observation means that half of the digested protein reaching blood is still bearing some information in the form of the AA chain. Much research has tried to investigate what might be the best source of protein for bone health.

#### **5.1 Animal vs. vegetable protein**

Some observational studies considered the nature of protein source when measuring the effect on bone. One of them studied the relationships between the animal/vegetable protein ratio and bone parameters on 1035 postmenopausal women, showing that the ratio is positively associated with bone loss and hip fracture risk (Sellmeyer *et al.*, 2001). However, the fact that a ratio was used in this study instead of the absolute values has been criticized (Bonjour, 2005; M. Thorpe & Evans, 2011). Other observational studies provide conflicting results. BMD was positively associated with animal protein intake and negatively to vegetable intake in one study (Promislow *et al.*, 2002) but the opposite association was found when considering bone ultrasound attenuation (Weikert *et al.*, 2005). Low levels of both protein types have been associated with deleterious effects on bone: low animal protein is related to bone loss (Hannan *et al.*, 2000) and low vegetable protein is related to low BMD (Beasley *et al.*, 2010). These results suggest that a minimum intake of both proteins is required regardless of the source. Finally, when considering fracture risk, a positive relationship was found with both animal (Dargent-Molina *et al.*, 2008; Feskanich *et al.*, 1996; Meyer *et al.*, 1997) and vegetable protein (Munger *et al.*, 1999; D. L. Thorpe *et al.*, 2008a). It should be noted that the studies finding a positive relationship between animal protein and fracture risk also found a positive relationship with total protein. On the other hand, those finding a positive relationship between vegetable protein and fracture risk reported a negative one with total protein. Hence the results for animal and vegetable protein are not obtained in the same context.

The intervention studies comparing animal and vegetable protein always focus on specific types of protein, usually soy and casein. Hence it is the specific effect of those proteins that is evaluated and not the one of the animal and vegetable food groups. The only whole-diet intervention study was based on meat or protein-rich vegetables such as nuts and grains. Only urinary parameters were measured and urinary calcium was similar for the two diets (Massey & Kynast-Gales, 2001). More studies of this type are needed to address this issue.

Mechanistically, the reason for differentiating animal from vegetable protein is their differing sulphur AA content and the consequent modulation of potential acid load (Sellmeyer *et al.*, 2001). The influence of sulphate content on protein efficiency has been emphasized in a cross-sectional study on 161 postmenopausal women. The results show a positive association of total protein intake with lumbar spine and total hip BMD; however, in lumbar spine this benefit is suppressed by the sulphur-containing AAs (M. Thorpe *et al.*, 2008b). The authors conclude that an excessive consumption of sulphur AAs is likely to be deleterious for bone. However, as underlined by Massey, the assumption that animal protein contains more sulphur AA than vegetable protein is not always true. As an example, potential acidity from sulphur AA in milk or beef is lower than that of whole wheat or white rice (Massey, 2003). Hence, even if sulphur AAs were proven to have a negative effect on bone, the extrapolation to animal and vegetable food groups is likely to give incorrect results because of the broad variety of foods in these groups. In future research, it would be more accurate to consider an estimation of the sulphate content of each protein-containing food rather than focusing only on a crude animal-vegetable distinction.

#### **5.2 Milk proteins**

128 Dual Energy X-Ray Absorptiometry

intestinal barrier, 50% are completely degraded as simple amino acids, 40% are partly degraded as peptides and 10% remain intact (Mallegol *et al.*, 2005). This observation means that half of the digested protein reaching blood is still bearing some information in the form of the AA chain. Much research has tried to investigate what might be the best source of

Some observational studies considered the nature of protein source when measuring the effect on bone. One of them studied the relationships between the animal/vegetable protein ratio and bone parameters on 1035 postmenopausal women, showing that the ratio is positively associated with bone loss and hip fracture risk (Sellmeyer *et al.*, 2001). However, the fact that a ratio was used in this study instead of the absolute values has been criticized (Bonjour, 2005; M. Thorpe & Evans, 2011). Other observational studies provide conflicting results. BMD was positively associated with animal protein intake and negatively to vegetable intake in one study (Promislow *et al.*, 2002) but the opposite association was found when considering bone ultrasound attenuation (Weikert *et al.*, 2005). Low levels of both protein types have been associated with deleterious effects on bone: low animal protein is related to bone loss (Hannan *et al.*, 2000) and low vegetable protein is related to low BMD (Beasley *et al.*, 2010). These results suggest that a minimum intake of both proteins is required regardless of the source. Finally, when considering fracture risk, a positive relationship was found with both animal (Dargent-Molina *et al.*, 2008; Feskanich *et al.*, 1996; Meyer *et al.*, 1997) and vegetable protein (Munger *et al.*, 1999; D. L. Thorpe *et al.*, 2008a). It should be noted that the studies finding a positive relationship between animal protein and fracture risk also found a positive relationship with total protein. On the other hand, those finding a positive relationship between vegetable protein and fracture risk reported a negative one with total protein. Hence the results for animal and vegetable protein are not

The intervention studies comparing animal and vegetable protein always focus on specific types of protein, usually soy and casein. Hence it is the specific effect of those proteins that is evaluated and not the one of the animal and vegetable food groups. The only whole-diet intervention study was based on meat or protein-rich vegetables such as nuts and grains. Only urinary parameters were measured and urinary calcium was similar for the two diets (Massey & Kynast-Gales, 2001). More studies of this type are needed to address this issue. Mechanistically, the reason for differentiating animal from vegetable protein is their differing sulphur AA content and the consequent modulation of potential acid load (Sellmeyer *et al.*, 2001). The influence of sulphate content on protein efficiency has been emphasized in a cross-sectional study on 161 postmenopausal women. The results show a positive association of total protein intake with lumbar spine and total hip BMD; however, in lumbar spine this benefit is suppressed by the sulphur-containing AAs (M. Thorpe *et al.*, 2008b). The authors conclude that an excessive consumption of sulphur AAs is likely to be deleterious for bone. However, as underlined by Massey, the assumption that animal protein contains more sulphur AA than vegetable protein is not always true. As an example, potential acidity from sulphur AA in milk or beef is lower than that of whole wheat or white rice (Massey, 2003). Hence, even if sulphur AAs were proven to have a negative effect on bone, the extrapolation to animal and vegetable food groups is likely to give incorrect

protein for bone health.

**5.1 Animal vs. vegetable protein** 

obtained in the same context.

Milk contains many bioactive factors, including growth hormones, enzymes, antimicrobials, anti-inflammatory agents, transporters and peptide or nonpeptide hormones. Milk, because it contains bioactive molecules, extends beyond applications in infant nutrition and was considered as a possible source of factors with anabolic effects on bone.

#### **5.2.1 Milk Basic Proteins (MBP)**

Clinical trials showed that MBP supplementation increased BMD and decreased bone resorption biomarkers in healthy women (Aoe *et al.*, 2001; Uenishi *et al.*, 2007; Yamamura *et al.*, 2002), menopausal women (Aoe *et al.*, 2005), and healthy older women (Aoyagi *et al.*, 2010). In particular, MBP suppresses osteoclast-mediated bone resorption and leads to reduced osteoclast number in animal studies (Morita *et al.*, 2008). Morita *et al*. reported that the protein fraction responsible for the observed activities of MBP is the bovine angiogenin which acts as a bone resorption-inhibitory protein (Morita *et al.*, 2008).

However, not all studies have shown a beneficial effect of MBP. In one study, 84 healthy young women were divided into three groups receiving placebo, whole milk, or milk containing 40 mg MBP for 8 months. Compared with the baseline values, total BMD significantly increased in all groups. There was a significant decrease of bone resorption marker N-teleopeptides of type-I collagen (NTx) while bone formation remained stable in both milk groups. (Zou *et al.*, 2009).

HPLC analysis of the major proteins in the MBP fraction identified the presence of the glycoprotein lactoferrin in most of these fractions (Naot *et al.*, 2005).

#### **5.2.2 Lactoferrin**

Lactoferrin (LF) is an 80 kDa iron-binding glycoprotein of the transferrin family. This molecule has been demonstrated to inhibit *in vitro* osteoclast-mediated bone resorption (Lorget *et al.*, 2002). LF was also demonstrated to have *in vitro* anabolic, differentiating and anti-apoptotic effects on osteoblasts, and to inhibit osteoclastogenesis. Moreover *in vivo* local injection of LF above the hemicalvaria increases bone formation and bone area in adult mice (Cornish *et al.*, 2004). LF has a role in host non-specific defense (Gahr *et al.*, 1991; Legrand *et al.*, 2004). In addition to its direct antimicrobial effects, LF is believed to modulate the inflammatory process mainly by preventing the release of inflammatory cytokines which induce recruitment and activation of immune cells at inflammatory sites (Legrand *et al.*, 2005).

Investigations of our group and others established that LF at physiological concentrations can stimulate proliferation of primary osteoblasts and osteoblastic-cell lines and increase osteoblast differentiation (Blais *et al.*, 2009; Cornish, 2004; Takayama & Mizumachi, 2008). Studies using 3-week culture of primary rat osteoblasts show that LF dose-dependently increases the number of nodules and the area of mineralized bone formed (Cornish, 2004). Our *in vitro* experiments demonstrated that LF could directly act on bone cells. Bovine LF

Dietary Protein and Bone Health 131

inhibition in order to improve bone loss (Malet *et al.*, 2011). Recently a clinical trial of 38 healthy postmenopausal women randomized into placebo or ribonuclease-enriched-LF (R-ELF) groups evaluated bone health status over a period of 12 months. The authors demonstrated that R-ELF supplementation reduced bone resorption and increased osteoblastic bone formation; however BMD was not evaluated (Bharadwaj *et al.*, 2009).

In conclusion, LF has a positive effect on bone health and might be useful in pathological states of reduced bone density. The molecular mechanisms are not fully understood but our studies suggest that dietary bLF supplementation can have a beneficial effect on postmenopausal bone loss not only via a direct effect but also by modulating immune function. The development of pharmaceutical or nutriceutical compounds that are based on

Collagen has a unique triple helix configuration with a repeating sequence (Gly-X-Y)n, with X and Y being mostly proline and hydroxyproline (Hyp) (Bos *et al.*, 1999; Ramshaw *et al.*, 1998). Some studies suggest that a hydrolyzed collagen-enriched diet improves bone collagen metabolism and BMD. Oral administration of Hydrolyzed Collagen (HC) increased BMC and BMD in rats and mice fed a calcium or protein deficient diet (Koyama *et al.*, 2001;

Oral administration of HC was demonstrated to increase the quantity of type I collagen and proteoglycans in the bone matrix of ovariectomized rats (Nomura *et al.*, 2005). Moreover, in patients with osteoporosis, oral intake of HC with calcitonin had a stronger inhibitory effect on bone resorption than calcitonin alone (Adam *et al.*, 1996). Oesser *et al*. demonstrated the intestinal absorption and cartilage accumulation of collagen-derived peptides (Oesser *et al.*, 1999). It has been generally assumed that collagen–rich diets interact with the bone matrix. Indeed, collagen-derived di- and tripeptides rich in hydroxyproline such as Hyp, Pro-Hyp, Pro-Hyp-Gly or Gly-Pro-Val have been detected in human blood following the ingestion of HC (Iwai *et al.*, 2005). The PEPT1 proton-dependent transporter assures the transport of Pro-

A study of Ohara *et al*. compared quantity and structures of food-derived gelatin hydrolysates in human blood from fish scale, fish skin and pork skin type I collagen in a single blind crossover study (Ohara *et al.*, 2007). Amounts of free Hyp and Hyp-containing peptide were measured over a 24-h period. Hyp-containing peptides comprised approximately 30% of all detected Hyp. However, efficiency of HC ingestion depends not only on collagen origin but also on the molecular size of the HC. Collagen needs to be

Our *in vivo* studies indicate that ingestion of HC diet induced the growth of the external diameter of the bone cortical zone in OVX mice (Guillerminet *et al.*, 2010, 2011). The increased cortical area was correlated with a significant increase in the femur external

LF will require a better understanding of LF's mechanism of action on bone.

**5.3 Collagen** 

Wu *et al.*, 2004).

**5.3.1 Ingestion of collagen** 

**5.3.2** *In vivo* **studies** 

Hyp across the intestinal barrier (Aito-Inoue *et al.*, 2007).

hydrolysed to be able to interact with bone metabolism.

(bLF) at low physiological concentrations (5 µg/ml) stimulates growth and activity of osteoblastic MC3T3-E1 cells and primary culture of murine osteoblast bone cells (Blais *et al.*, 2009). Low density lipoprotein receptor-related protein 1 and 2, which are present on osteoblastic cells, have been shown to be partially responsible for LF's mitogenic effect in osteoblasts (Grey *et al.*, 2004). Moreover, Grey *et al*., showed that LF is able to protect osteoblastic cells from apoptosis induced by serum withdrawal (Grey *et al.*, 2006).

LF action on osteoclasts is strikingly different since it produces an important arrest of osteoclastogenesis (Cornish, 2004; Lorget *et al.*, 2002). However, LF does not modulate mature ostoeclast activity. bLF at a concentration ranging from 10 to 1000 µg/ml was found to inhibit pre-osteoclastic established RAW cell growth. These results were confirmed in mixed primary culture of murine bone cells (Blais *et al.*, 2009). In contrast, there was no effect of LF on bone resorption when tested on isolated mature osteoclasts, or in organ cultures that can detect mature osteoclast activity (Grey *et al.*, 2006).

*In vivo* bone effects of LF were first studied using local injection of LF above the hemicalvaria which increased bone formation and bone area in adult mice (Cornish, 2004). Few recent studies using ovariectomized (OVX) rodents as a model for post menopausal bone loss measured the effect of dietary supplementation on bone (Blais *et al.*, 2009; Guo *et al.*, 2009; Malet *et al.*, 2011). LF administered orally to OVX rats for three months protected them against the OVX-induced reduction of bone volume and BMD and increased the parameters of mechanical strength, increased bone formation and reduced bone resorption (Guo *et al.*, 2009). Our studies using OVX mice demonstrated that the dietary bLF transfer into peripheral blood. LF supplementation increases BMD and bone strength compared to the OVX control group. This study supports a direct effect of LF on bone cells (Blais *et al.*, 2009).

Recent animal studies demonstrated that estrogen deficiency causes bone loss by mechanisms associated with inflammatory and oxidative processes (Grassi *et al.*, 2007; Lean *et al.*, 2003; Muthusami *et al.*, 2005). TNFis one of the cytokines responsible for the augmented osteoclastogenesis (Suda *et al.*, 1999). Indeed, ovariectomy causes an increase in TNF production from T-cells which in turn increases macrophage colony-stimulating factor and RANKL levels, leading to osteoclastogenesis (Cornish *et al.*, 2004; Suda *et al.*, 1999). Moreover the presence of increased levels of TNFwas reported in the bone marrow of OVX animals and in blood cells of postmenopausal women (Oh *et al.*, 2007; Shanker *et al.*, 1994). Postmenopausal osteoporosis should also be regarded as the result of an inflammatory process (Weitzmann & Pacifici, 2007). It has been shown that bLF plays a role in host non-specific defense and modulates the inflammatory process mainly by preventing the release of inflammatory cytokines that induce recruitment and activation of immune cells at inflammatory sites (Debbabi *et al.*, 1998; Legrand *et al.*, 2006). Indeed, bLF enriched diet ingestion can reduce release of pro-inflammatory cytokines and increase anti-inflammatory cytokine production. Our studies showed that bLF ingestion decreases bone loss and bone resorption markers. A decreased TNFmRNA expression associated with a TNFproduction inhibition on peripheral T-lymphocytes was observed with a bLF supplementation in OVX mice. Furthermore, bLF can prevent lymphocyte activation and cytokine release in the bone micro-environment. Production and release of TNFwere strongly down-regulated by LF. These immune modulations were spatially and temporally correlated with reduced bone loss. We suggested that bLF modulates the inflammatory process via specific TNF

(bLF) at low physiological concentrations (5 µg/ml) stimulates growth and activity of osteoblastic MC3T3-E1 cells and primary culture of murine osteoblast bone cells (Blais *et al.*, 2009). Low density lipoprotein receptor-related protein 1 and 2, which are present on osteoblastic cells, have been shown to be partially responsible for LF's mitogenic effect in osteoblasts (Grey *et al.*, 2004). Moreover, Grey *et al*., showed that LF is able to protect

LF action on osteoclasts is strikingly different since it produces an important arrest of osteoclastogenesis (Cornish, 2004; Lorget *et al.*, 2002). However, LF does not modulate mature ostoeclast activity. bLF at a concentration ranging from 10 to 1000 µg/ml was found to inhibit pre-osteoclastic established RAW cell growth. These results were confirmed in mixed primary culture of murine bone cells (Blais *et al.*, 2009). In contrast, there was no effect of LF on bone resorption when tested on isolated mature osteoclasts, or in organ

*In vivo* bone effects of LF were first studied using local injection of LF above the hemicalvaria which increased bone formation and bone area in adult mice (Cornish, 2004). Few recent studies using ovariectomized (OVX) rodents as a model for post menopausal bone loss measured the effect of dietary supplementation on bone (Blais *et al.*, 2009; Guo *et al.*, 2009; Malet *et al.*, 2011). LF administered orally to OVX rats for three months protected them against the OVX-induced reduction of bone volume and BMD and increased the parameters of mechanical strength, increased bone formation and reduced bone resorption (Guo *et al.*, 2009). Our studies using OVX mice demonstrated that the dietary bLF transfer into peripheral blood. LF supplementation increases BMD and bone strength compared to the OVX control group. This study supports a direct effect of LF on bone cells (Blais *et al.*,

Recent animal studies demonstrated that estrogen deficiency causes bone loss by mechanisms associated with inflammatory and oxidative processes (Grassi *et al.*, 2007; Lean *et al.*, 2003; Muthusami *et al.*, 2005). TNFis one of the cytokines responsible for the augmented osteoclastogenesis (Suda *et al.*, 1999). Indeed, ovariectomy causes an increase in TNF production from T-cells which in turn increases macrophage colony-stimulating factor and RANKL levels, leading to osteoclastogenesis (Cornish *et al.*, 2004; Suda *et al.*, 1999). Moreover the presence of increased levels of TNFwas reported in the bone marrow of OVX animals and in blood cells of postmenopausal women (Oh *et al.*, 2007; Shanker *et al.*, 1994). Postmenopausal osteoporosis should also be regarded as the result of an inflammatory process (Weitzmann & Pacifici, 2007). It has been shown that bLF plays a role in host non-specific defense and modulates the inflammatory process mainly by preventing the release of inflammatory cytokines that induce recruitment and activation of immune cells at inflammatory sites (Debbabi *et al.*, 1998; Legrand *et al.*, 2006). Indeed, bLF enriched diet ingestion can reduce release of pro-inflammatory cytokines and increase anti-inflammatory cytokine production. Our studies showed that bLF ingestion decreases bone loss and bone resorption markers. A decreased TNFmRNA expression associated with a TNFproduction inhibition on peripheral T-lymphocytes was observed with a bLF supplementation in OVX mice. Furthermore, bLF can prevent lymphocyte activation and cytokine release in the bone micro-environment. Production and release of TNFwere strongly down-regulated by LF. These immune modulations were spatially and temporally correlated with reduced bone loss. We suggested that bLF modulates the inflammatory process via specific TNF

osteoblastic cells from apoptosis induced by serum withdrawal (Grey *et al.*, 2006).

cultures that can detect mature osteoclast activity (Grey *et al.*, 2006).

2009).

inhibition in order to improve bone loss (Malet *et al.*, 2011). Recently a clinical trial of 38 healthy postmenopausal women randomized into placebo or ribonuclease-enriched-LF (R-ELF) groups evaluated bone health status over a period of 12 months. The authors demonstrated that R-ELF supplementation reduced bone resorption and increased osteoblastic bone formation; however BMD was not evaluated (Bharadwaj *et al.*, 2009).

In conclusion, LF has a positive effect on bone health and might be useful in pathological states of reduced bone density. The molecular mechanisms are not fully understood but our studies suggest that dietary bLF supplementation can have a beneficial effect on postmenopausal bone loss not only via a direct effect but also by modulating immune function. The development of pharmaceutical or nutriceutical compounds that are based on LF will require a better understanding of LF's mechanism of action on bone.

#### **5.3 Collagen**

Collagen has a unique triple helix configuration with a repeating sequence (Gly-X-Y)n, with X and Y being mostly proline and hydroxyproline (Hyp) (Bos *et al.*, 1999; Ramshaw *et al.*, 1998). Some studies suggest that a hydrolyzed collagen-enriched diet improves bone collagen metabolism and BMD. Oral administration of Hydrolyzed Collagen (HC) increased BMC and BMD in rats and mice fed a calcium or protein deficient diet (Koyama *et al.*, 2001; Wu *et al.*, 2004).

#### **5.3.1 Ingestion of collagen**

Oral administration of HC was demonstrated to increase the quantity of type I collagen and proteoglycans in the bone matrix of ovariectomized rats (Nomura *et al.*, 2005). Moreover, in patients with osteoporosis, oral intake of HC with calcitonin had a stronger inhibitory effect on bone resorption than calcitonin alone (Adam *et al.*, 1996). Oesser *et al*. demonstrated the intestinal absorption and cartilage accumulation of collagen-derived peptides (Oesser *et al.*, 1999). It has been generally assumed that collagen–rich diets interact with the bone matrix. Indeed, collagen-derived di- and tripeptides rich in hydroxyproline such as Hyp, Pro-Hyp, Pro-Hyp-Gly or Gly-Pro-Val have been detected in human blood following the ingestion of HC (Iwai *et al.*, 2005). The PEPT1 proton-dependent transporter assures the transport of Pro-Hyp across the intestinal barrier (Aito-Inoue *et al.*, 2007).

A study of Ohara *et al*. compared quantity and structures of food-derived gelatin hydrolysates in human blood from fish scale, fish skin and pork skin type I collagen in a single blind crossover study (Ohara *et al.*, 2007). Amounts of free Hyp and Hyp-containing peptide were measured over a 24-h period. Hyp-containing peptides comprised approximately 30% of all detected Hyp. However, efficiency of HC ingestion depends not only on collagen origin but also on the molecular size of the HC. Collagen needs to be hydrolysed to be able to interact with bone metabolism.

#### **5.3.2** *In vivo* **studies**

Our *in vivo* studies indicate that ingestion of HC diet induced the growth of the external diameter of the bone cortical zone in OVX mice (Guillerminet *et al.*, 2010, 2011). The increased cortical area was correlated with a significant increase in the femur external

Dietary Protein and Bone Health 133

Taken together, the results indicate that hydrolyzed collagen modulates bone formation and mineralization of the bone matrix by stimulating osteoblast growth and differentiation while reducing osteoclast differentiation. These effects lead to growth of the external diameter of

Soy contains isoflavones able to bind to estrogen receptors (Folman & Pope, 1969). They have received considerable interest as a possible alternative to conventional Hormone Replacement Therapy (HRT). However, the efficiency of phytoestrogens such as soy

Epidemiological studies suggest that populations with high soy intake (such as Asian populations) have a lower incidence of osteoporotic fractures (Adlercreutz & Mazur, 1997; Schwartz *et al.*, 1999). Asian women typically consume about 20g of soy daily which provides 40 mg of isoflavones (Chen *et al.*, 1999; Ho *et al.*, 2003). However, lower rates of fracture in these populations may not be fully attributed to soy consumption as ethnic related variation in fracture rates can also be explained by differences in bone structure

Many animal studies show that soy protein and/or its isoflavones have positive effects on bone mineral density (BMD) (Arjmandi *et al.*, 1998a, 1998b). However, clinical trial results ranged from no significant changes (Alekel *et al.*, 2000; Dalais *et al.*, 1998; Gallagher *et al.*, 2004; Kreijkamp-Kaspers *et al.*, 2004; Potter *et al.*, 1998) to a slight increase (Chiechi *et al.*, 2002; Lydeking-Olsen *et al.*, 2004; Potter *et al.*, 1998) in BMD. The bone protective effects of

The major isoflavones in soy foods include genistein and diadzein. Genistein 2 has one-third of the potency of estradiol 1 when it interacts with estrogen receptor-b (ER-b), and onethousandth of the potency of estradiol 1 when it interacts with ER-a. Hence Genistein 2 can induce a small estradiol-like response in bone tissues (Adlercreutz & Mazur, 1997; Zhou *et al.*, 2003). Another isoflavone, called equol, is not present in soybean but is a metabolic product of the biotransformation of diadzein by gut bacteria (Setchell *et al.*, 2002). 80% of the Asian population are equol producers (Fujimoto *et al.*, 2008; Morton *et al.*, 2002). In contrast, as few as 25% of individuals in North America and Europe are able to make S-equol (Lampe

A number of reviews describe the effects of dietary soy and isoflavones on bone (Jackson *et al.*, 2011; Messina, 2010; Reinwald & Weaver, 2010). Among the studies exploring the effect of isoflavone-containing food on BMD in postmenopausal women, few report a relationship between soy consumption and the risk of bone fracture. A clinical trial conducted by Marini *et al*. found that in postmenopausal osteopenic Italians receiving 54mg/day genistein for two years, spinal BMD increased by 5.8% (n=150), whereas it decreased in the placebo group by 6.3% (n=154). Similar effects were reported for the hip (Marini *et al.*, 2007). However,

the cortical zone.

(Bouxsein, 2011).

*et al.*, 1998).

**5.4.1 Types of isoflavones** 

**5.4.2 Isoflavone and bone fracture** 

**5.4 Isoflavone-containing soy protein** 

isoflavone on bone is still to be proven.

soy and/or its isoflavones are at best inconclusive.

diameter, without modification of the size of the medullar area. Therefore, the increased size of the cortical area was induced by a periosteal apposition of bone on the mouse femur. Due to this increase in bone size, the ultimate strength of femurs of OVX-mice ingesting HC was significantly greater than the control OVX mice. The increase of the external diameter was also related to a higher level of bone ALP during the first month of HC ingestion. However, the effect was transient; after three months no significant ALP increase was reported. Moreover, HC ingestion was able to increase the bone non-mineral content. There was no significant modification of Young's modulus but bone stiffness increased. Assuming that the stiffness of bone is correlated to the amount of type I collagen present (Burr, 2002; Mann *et al.*, 2001), and since some previous studies showed an increase of type I collagen and proteoglycan excretion for mice fed hydrolyzed collagen, we can propose that HC ingestion increases type I collagen formation in mouse bone.

#### **5.3.3** *In vitro* **studies**

The *in vitro* results obtained with primary tissue culture of murine bone cells confirmed that HC was able to stimulate cell growth and ALP activity. In our studies, all the tested collagens were able to increase osteoblast activity but the 2kDA porcine HC was the most efficient *in vitro* (Guillerminet *et al.*, 2010). Similar observations were also reported with osteoblasts grown on collagen type I films compared to a plastic support with an improvement in various bone markers including increased ALP activity and an accelerated and uniform mineralization of the bone matrix (Lynch *et al.*, 1995). Moreover, our work using the *in vitro* BD BioCoatTM OsteologicTM bone cell culture system showed that PCH-N hydrolyzed collagen did not modify osteoclast growth but reduced osteoclast differentiation (Guillerminet *et al.*, 2010). This effect, combined with increased osteoblast activity is likely to modulate bone turnover leading to the growth of the external diameter of cortical bone.

#### **5.3.4 Mechanism of action**

Several potential mechanisms can be proposed to explain the influence of HC-derived peptides on bone metabolism. Some results have suggested that ingestion of type I hydrolyzed collagen leads to the production and absorption of collagen-derived peptides similar to peptides released from type I collagen *in situ* during bone resorption. Those peptides also act on bone cell metabolism (Adam *et al.*, 1996). Osteoblast activity involves three steps including proliferation, matrix protein synthesis (type I collagen and proteoglycans) and mineralization of the bone matrix (Owen *et al.*, 1991; Quarles *et al.*, 1992; Stein & Lian, 1993). Several hormones and cytokines can modulate osteoblast and osteoclast differentiation and activity. The cytokine TGF-β which is stored in a latent form in the bone matrix, and secreted during the bone resorption phase, is believed to exert such an effect (Oreffo *et al.*, 1989). TGF-β stimulates type I collagen and proteoglycan production while inhibiting that of hydroxyapatite. Interestingly, the type I collagen-derived peptide DGEA (asparagine, glycine, glutamine and alanine), was shown to interact with α2β1 integrin located on the osteoblast cell membrane. This interaction leads to inhibition of TGF-β and consequently bone matrix protein synthesis (Oesser *et al.*, 1999; Takeuchi *et al.*, 1996, 1997; Xiao *et al.*, 1998). Moreover, Hyp is an aromatic AA, and an increase of its concentration can, as suggested previously, increase IGF-1 levels which consequently attenuates bone loss.

diameter, without modification of the size of the medullar area. Therefore, the increased size of the cortical area was induced by a periosteal apposition of bone on the mouse femur. Due to this increase in bone size, the ultimate strength of femurs of OVX-mice ingesting HC was significantly greater than the control OVX mice. The increase of the external diameter was also related to a higher level of bone ALP during the first month of HC ingestion. However, the effect was transient; after three months no significant ALP increase was reported. Moreover, HC ingestion was able to increase the bone non-mineral content. There was no significant modification of Young's modulus but bone stiffness increased. Assuming that the stiffness of bone is correlated to the amount of type I collagen present (Burr, 2002; Mann *et al.*, 2001), and since some previous studies showed an increase of type I collagen and proteoglycan excretion for mice fed hydrolyzed collagen, we can propose that HC ingestion

The *in vitro* results obtained with primary tissue culture of murine bone cells confirmed that HC was able to stimulate cell growth and ALP activity. In our studies, all the tested collagens were able to increase osteoblast activity but the 2kDA porcine HC was the most efficient *in vitro* (Guillerminet *et al.*, 2010). Similar observations were also reported with osteoblasts grown on collagen type I films compared to a plastic support with an improvement in various bone markers including increased ALP activity and an accelerated and uniform mineralization of the bone matrix (Lynch *et al.*, 1995). Moreover, our work using the *in vitro* BD BioCoatTM OsteologicTM bone cell culture system showed that PCH-N hydrolyzed collagen did not modify osteoclast growth but reduced osteoclast differentiation (Guillerminet *et al.*, 2010). This effect, combined with increased osteoblast activity is likely to modulate bone turnover leading to the growth of the external diameter of cortical bone.

Several potential mechanisms can be proposed to explain the influence of HC-derived peptides on bone metabolism. Some results have suggested that ingestion of type I hydrolyzed collagen leads to the production and absorption of collagen-derived peptides similar to peptides released from type I collagen *in situ* during bone resorption. Those peptides also act on bone cell metabolism (Adam *et al.*, 1996). Osteoblast activity involves three steps including proliferation, matrix protein synthesis (type I collagen and proteoglycans) and mineralization of the bone matrix (Owen *et al.*, 1991; Quarles *et al.*, 1992; Stein & Lian, 1993). Several hormones and cytokines can modulate osteoblast and osteoclast differentiation and activity. The cytokine TGF-β which is stored in a latent form in the bone matrix, and secreted during the bone resorption phase, is believed to exert such an effect (Oreffo *et al.*, 1989). TGF-β stimulates type I collagen and proteoglycan production while inhibiting that of hydroxyapatite. Interestingly, the type I collagen-derived peptide DGEA (asparagine, glycine, glutamine and alanine), was shown to interact with α2β1 integrin located on the osteoblast cell membrane. This interaction leads to inhibition of TGF-β and consequently bone matrix protein synthesis (Oesser *et al.*, 1999; Takeuchi *et al.*, 1996, 1997; Xiao *et al.*, 1998). Moreover, Hyp is an aromatic AA, and an increase of its concentration can, as suggested previously, increase IGF-1 levels which consequently attenuates bone loss.

increases type I collagen formation in mouse bone.

**5.3.3** *In vitro* **studies** 

**5.3.4 Mechanism of action** 

Taken together, the results indicate that hydrolyzed collagen modulates bone formation and mineralization of the bone matrix by stimulating osteoblast growth and differentiation while reducing osteoclast differentiation. These effects lead to growth of the external diameter of the cortical zone.

#### **5.4 Isoflavone-containing soy protein**

Soy contains isoflavones able to bind to estrogen receptors (Folman & Pope, 1969). They have received considerable interest as a possible alternative to conventional Hormone Replacement Therapy (HRT). However, the efficiency of phytoestrogens such as soy isoflavone on bone is still to be proven.

Epidemiological studies suggest that populations with high soy intake (such as Asian populations) have a lower incidence of osteoporotic fractures (Adlercreutz & Mazur, 1997; Schwartz *et al.*, 1999). Asian women typically consume about 20g of soy daily which provides 40 mg of isoflavones (Chen *et al.*, 1999; Ho *et al.*, 2003). However, lower rates of fracture in these populations may not be fully attributed to soy consumption as ethnic related variation in fracture rates can also be explained by differences in bone structure (Bouxsein, 2011).

Many animal studies show that soy protein and/or its isoflavones have positive effects on bone mineral density (BMD) (Arjmandi *et al.*, 1998a, 1998b). However, clinical trial results ranged from no significant changes (Alekel *et al.*, 2000; Dalais *et al.*, 1998; Gallagher *et al.*, 2004; Kreijkamp-Kaspers *et al.*, 2004; Potter *et al.*, 1998) to a slight increase (Chiechi *et al.*, 2002; Lydeking-Olsen *et al.*, 2004; Potter *et al.*, 1998) in BMD. The bone protective effects of soy and/or its isoflavones are at best inconclusive.

#### **5.4.1 Types of isoflavones**

The major isoflavones in soy foods include genistein and diadzein. Genistein 2 has one-third of the potency of estradiol 1 when it interacts with estrogen receptor-b (ER-b), and onethousandth of the potency of estradiol 1 when it interacts with ER-a. Hence Genistein 2 can induce a small estradiol-like response in bone tissues (Adlercreutz & Mazur, 1997; Zhou *et al.*, 2003). Another isoflavone, called equol, is not present in soybean but is a metabolic product of the biotransformation of diadzein by gut bacteria (Setchell *et al.*, 2002). 80% of the Asian population are equol producers (Fujimoto *et al.*, 2008; Morton *et al.*, 2002). In contrast, as few as 25% of individuals in North America and Europe are able to make S-equol (Lampe *et al.*, 1998).

#### **5.4.2 Isoflavone and bone fracture**

A number of reviews describe the effects of dietary soy and isoflavones on bone (Jackson *et al.*, 2011; Messina, 2010; Reinwald & Weaver, 2010). Among the studies exploring the effect of isoflavone-containing food on BMD in postmenopausal women, few report a relationship between soy consumption and the risk of bone fracture. A clinical trial conducted by Marini *et al*. found that in postmenopausal osteopenic Italians receiving 54mg/day genistein for two years, spinal BMD increased by 5.8% (n=150), whereas it decreased in the placebo group by 6.3% (n=154). Similar effects were reported for the hip (Marini *et al.*, 2007). However,

Dietary Protein and Bone Health 135

protein-induced IGF-1 without the negative effect of the acid load by compensating the diet

Dietary protein quality adds complexity to the protein debate. It has been hypothesized that animal protein would be more deleterious to bone than vegetal protein. However, studies show no real difference between those two protein sources. Similarly, long-term observational studies support a benefit of traditional soy food consumption on bone health, but no conclusive evidence supports the hypothesis that this is due to the isoflavones. On the other hand, some peptides obtained from protein digestion have been shown to be helpful to prevent bone loss. Recent results indicate that HC could be of potential interest for nutritional intervention in the prevention of bone loss. Moreover, LF has been reported to have a positive effect on bone health and might be useful in pathological states of reduced bone density. The molecular mechanisms are not fully understood but our studies suggest that dietary bLF supplementation can have a beneficial effect on postmenopausal bone loss

Adam, M.; Spacek, P.; Hulejova, H.; Galianova, A. & Blahos, J. (1996). [Postmenopausal

Adlercreutz, H. & Mazur, W. (1997). Phyto-oestrogens and Western diseases. *Ann Med*, Vol.

Aito-Inoue, M.; Lackeyram, D.; Fan, M.Z.; Sato, K. & Mine, Y. (2007). Transport of a

Alekel, D.L.; Germain, A.S.; Peterson, C.T.; Hanson, K.B.; Stewart, J.W. & Toda, T. (2000).

Alexy, U.; Remer, T.; Manz, F.; Neu, C.M. & Schoenau, E. (2005). Long-term protein intake

Aoe, S.; Koyama, T.; Toba, Y.; Itabashi, A. & Takada, Y. (2005). A controlled trial of the effect

Aoyagi, Y.; Park, H.; Park, S.; Yoshiuchi, K.; Kikuchi, H.; Kawakami, H.; Morita, Y.; Ono, A.

controlled trial. *Int Dairy J*, Vol. 20, No. (Mar 2010), pp. 724-730

osteoporosis. Treatment with calcitonin and a diet rich in collagen proteins]. *Cas Lek* 

tripeptide, Gly-Pro-Hyp, across the porcine intestinal brush-border membrane. *J* 

Isoflavone-rich soy protein isolate attenuates bone loss in the lumbar spine of perimenopausal women. *Am J Clin Nutr*, Vol. 72, No. 3, (Sep 2000), pp. 844-852 Alekel, D.L.; Van Loan, M.D.; Koehler, K.J.; Hanson, L.N.; Stewart, J.W.; Hanson, K.B.;

Kurzer, M.S. & Peterson, C.T. (2009). The soy isoflavones for reducing bone loss (SIRBL) study: a 3-y randomized controlled trial in postmenopausal women. *Am J* 

and dietary potential renal acid load are associated with bone modeling and remodeling at the proximal radius in healthy children. *Am J Clin Nutr*, Vol. 82, No.

of milk basic protein (MBP) supplementation on bone metabolism in healthy menopausal women. *Osteoporos Int*, Vol. 16, No. 12, (Dec 2005), pp. 2123-2128 Aoe, S.; Toba, Y.; Yamamura, J.; Kawakami, H.; Yahiro, M.; Kumegawa, M.; Itabashi, A. &

Takada, Y. (2001). Controlled trial of the effects of milk basic protein (MBP) supplementation on bone metabolism in healthy adult women. *Biosci Biotechnol* 

& Shephard RJ (2010). Interactive effects of milk basic protein supplements and habitual physical activity on bone health in older women: A 1-year randomized

not only by acting on bone cells but also by modulating immune function.

*Cesk*, Vol. 135, No. 3, (Jan 31 1996), pp. 74-78

*Pept Sci*, Vol. 13, No. 7, (Jul 2007), pp. 468-474

*Clin Nutr*, Vol. 91, No. 1, (Jan 2009), pp. 218-230

*Biochem*, Vol. 65, No. 4, (Apr 2001), pp. 913-918

29, No. 2, (Apr 1997), pp. 95-120

5, (Nov 2005), pp. 1107-1114

with adequate alkalinizing foods.

**7. References** 

recently published long-term trials do not confirm these results; only the trial conducted by Alekel *et al*. reports a modest effect at the femoral neck with 120mg/d isoflavone but no effect with 80mg/d (Alekel *et al.*, 2009).

In contrast clinical trials investigating associations between soy-food intake and BMD in Japanese or Chinese healthy postmenopausal women report that higher isoflavone consumption is associated with lower risk of bone fracture (Ho *et al.*, 2003; Ikeda *et al.*, 2006; Kaneki *et al.*, 2001). Analysis of fracture incidence in the Shanghai cohort (Zhang *et al.*, 2005) and of hip fracture in the Singapore cohort (Koh *et al.*, 2009) shows in both studies one-third reductions in fracture risk when comparing high- with low-soy consumers.

#### **5.4.3 Factors modulating the effect of soy on bone health**

The effectiveness of dietary adaptation of western populations which rarely consumed soy must be considered. East Asian participants in epidemiological studies did not require an adaptation period or an interruption of life-long dietary habits like a western population would. Hence the observation cannot be extrapolated from one population to another.

It may be less difficult to determine bone effects following a life-long intake of traditional foods compared with intermittent intakes of soy. Traditional soy foods are a complex blend of isoflavones, protein, lipids, vitamins, minerals and other bioactive compounds that may act individually and/or synergistically to exert healthy effects. Supplements included in western diets provide quantities of individual soy components. Types of whole soy food consumed (fermented vs. nonfermented) and/or ethnicity (equol producers) may also affect outcome interpretation of soy bone effects.

Long-term observational studies in Asian populations support a benefit of traditional soy food consumption on bone health in this population. The health effects of soy-bean phytoestrogens in non-Asian postmenopausal women are promising. No conclusive evidence supports that the isoflavones from the sources studied do have beneficial effects on bone health. More researches are needed to clarify the role of dietary phytoestrogen in osteoporosis prevention.

#### **6. Conclusion**

Protein acts on bone metabolism at different levels and through different mechanisms. There is little evidence that a high-protein diet will increase bone loss. Protein is well-known to be calciuric, yet there are conflicting data on whether the excreted calcium comes from an increase of calcium absorption or from bone resorption. The direct effects of protein on bone turnover markers and BMD seem to be positive when considering observational studies, but interventional studies do not provide significant outcomes to conclude. Finally, when considering fracture rate, there seems to be a small positive effect of protein on bone as long as calcium levels remain adequate.

Two mechanisms are proposed to explain the action of protein on bone: the acid-ash theory and the hormonal anabolic effect through IGF-1 and CaSR. The hormonal anabolic mechanism supports the fact that protein is beneficial to bone by increasing IGF-1. On the other hand, the acid-ash theory considers that the acid load due to protein consumption is harmful to bone. If both mechanisms occur at the same time, it is possible to benefit from the

recently published long-term trials do not confirm these results; only the trial conducted by Alekel *et al*. reports a modest effect at the femoral neck with 120mg/d isoflavone but no

In contrast clinical trials investigating associations between soy-food intake and BMD in Japanese or Chinese healthy postmenopausal women report that higher isoflavone consumption is associated with lower risk of bone fracture (Ho *et al.*, 2003; Ikeda *et al.*, 2006; Kaneki *et al.*, 2001). Analysis of fracture incidence in the Shanghai cohort (Zhang *et al.*, 2005) and of hip fracture in the Singapore cohort (Koh *et al.*, 2009) shows in both studies one-third

The effectiveness of dietary adaptation of western populations which rarely consumed soy must be considered. East Asian participants in epidemiological studies did not require an adaptation period or an interruption of life-long dietary habits like a western population would. Hence the observation cannot be extrapolated from one population to another.

It may be less difficult to determine bone effects following a life-long intake of traditional foods compared with intermittent intakes of soy. Traditional soy foods are a complex blend of isoflavones, protein, lipids, vitamins, minerals and other bioactive compounds that may act individually and/or synergistically to exert healthy effects. Supplements included in western diets provide quantities of individual soy components. Types of whole soy food consumed (fermented vs. nonfermented) and/or ethnicity (equol producers) may also affect

Long-term observational studies in Asian populations support a benefit of traditional soy food consumption on bone health in this population. The health effects of soy-bean phytoestrogens in non-Asian postmenopausal women are promising. No conclusive evidence supports that the isoflavones from the sources studied do have beneficial effects on bone health. More researches are needed to clarify the role of dietary phytoestrogen in

Protein acts on bone metabolism at different levels and through different mechanisms. There is little evidence that a high-protein diet will increase bone loss. Protein is well-known to be calciuric, yet there are conflicting data on whether the excreted calcium comes from an increase of calcium absorption or from bone resorption. The direct effects of protein on bone turnover markers and BMD seem to be positive when considering observational studies, but interventional studies do not provide significant outcomes to conclude. Finally, when considering fracture rate, there seems to be a small positive effect of protein on bone as long

Two mechanisms are proposed to explain the action of protein on bone: the acid-ash theory and the hormonal anabolic effect through IGF-1 and CaSR. The hormonal anabolic mechanism supports the fact that protein is beneficial to bone by increasing IGF-1. On the other hand, the acid-ash theory considers that the acid load due to protein consumption is harmful to bone. If both mechanisms occur at the same time, it is possible to benefit from the

reductions in fracture risk when comparing high- with low-soy consumers.

**5.4.3 Factors modulating the effect of soy on bone health** 

effect with 80mg/d (Alekel *et al.*, 2009).

outcome interpretation of soy bone effects.

osteoporosis prevention.

as calcium levels remain adequate.

**6. Conclusion** 

protein-induced IGF-1 without the negative effect of the acid load by compensating the diet with adequate alkalinizing foods.

Dietary protein quality adds complexity to the protein debate. It has been hypothesized that animal protein would be more deleterious to bone than vegetal protein. However, studies show no real difference between those two protein sources. Similarly, long-term observational studies support a benefit of traditional soy food consumption on bone health, but no conclusive evidence supports the hypothesis that this is due to the isoflavones. On the other hand, some peptides obtained from protein digestion have been shown to be helpful to prevent bone loss. Recent results indicate that HC could be of potential interest for nutritional intervention in the prevention of bone loss. Moreover, LF has been reported to have a positive effect on bone health and might be useful in pathological states of reduced bone density. The molecular mechanisms are not fully understood but our studies suggest that dietary bLF supplementation can have a beneficial effect on postmenopausal bone loss not only by acting on bone cells but also by modulating immune function.

#### **7. References**


Dietary Protein and Bone Health 137

Cao, J.J.; Johnson, L.K. & Hunt, J.R. (2011). A diet high in meat protein and potential renal

Ceglia, L.; Harris, S.S.; Abrams, S.A.; Rasmussen, H.M.; Dallal, G.E. & Dawson-Hughes, B.

Chen, Z.; Zheng, W.; Custer, L.J.; Dai, Q.; Shu, X.O.; Jin, F. & Franke, A.A. (1999). Usual

Chevalley, T.; Bonjour, J.P.; Ferrari, S. & Rizzoli, R. (2008). High-protein intake enhances the

Chevalley, T.; Hoffmeyer, P.; Bonjour, J.P. & Rizzoli, R. (2010). Early serum IGF-I response to

Chiechi, L.M.; Secreto, G.; D'Amore, M.; Fanelli, M.; Venturelli, E.; Cantatore, F.; Valerio, T.;

Conigrave, A.D. & Brown, E.M. (2006). Taste receptors in the gastrointestinal tract. II. L-

Conigrave, A.D.; Quinn, S.J. & Brown, E.M. (2000). L-amino acid sensing by the extracellular

Cornish, J. (2004). Lactoferrin promotes bone growth. *Biometals*, Vol. 17, No. 3, (Jun 2004),

Cornish, J.; Callon, K.E.; Naot, D.; Palmano, K.P.; Banovic, T.; Bava, U.; Watson, M.; Lin,

Currey, J. (2001). Sacrificial bonds heal bone. *Nature*, Vol. 414, No. 6865, (Dec 13 2001), pp. 699 Dalais, F.S.; Rice, G.E.; Wahlqvist, M.L.; Grehan, M.; Murkies, A.L.; Medley, G.; Ayton, R. &

Dargent-Molina, P.; Sabia, S.; Touvier, M.; Kesse, E.; Breart, G.; Clavel-Chapelon, F. &

homeostasis. *Annu Rev Nutr*, Vol. 28, No. 2008), pp. 131-155

276, No. 6, (Feb 9 2001), pp. 4070-4079

*Cancer*, Vol. 33, No. 1, 1999), pp. 82-87

Vol. 23, No. 1, (Jan 2008), pp. 131-142

Vol. 29, No. 1, (Feb 2010), pp. 78-83

Vol. 145, No. 9, (Sep 2004), pp. 4366-4374

*Climacteric*, Vol. 1, No. 2, (Jun 1998), pp. 124-129

*Miner Res*, Vol. 23, No. 12, (Dec 2008), pp. 1915-1922

4, (Aug 30 2002), pp. 295-300

4814-4819

pp. 331-335

women. *J Nutr*, Vol. 141, No. 3, (Mar 2011), pp. 391-397

*Clin Endocrinol Metab*, Vol. 94, No. 2, (Feb 2009), pp. 645-653

intracellular calcium mobilization and stimulation of bile flow. *J Biol Chem*, Vol.

acid load increases fractional calcium absorption and urinary calcium excretion without affecting markers of bone resorption or formation in postmenopausal

(2009). Potassium bicarbonate attenuates the urinary nitrogen excretion that accompanies an increase in dietary protein and may promote calcium absorption. *J* 

dietary consumption of soy foods and its correlation with the excretion rate of isoflavonoids in overnight urine samples among Chinese women in Shanghai. *Nutr* 

positive impact of physical activity on BMC in prepubertal boys. *J Bone Miner Res*,

oral protein supplements in elderly women with a recent hip fracture. *Clin Nutr*,

Laselva, G. & Loizzi, P. (2002). Efficacy of a soy rich diet in preventing postmenopausal osteoporosis: the Menfis randomized trial. *Maturitas*, Vol. 42, No.

amino acid sensing by calcium-sensing receptors: implications for GI physiology. *Am J Physiol Gastrointest Liver Physiol*, Vol. 291, No. 5, (Nov 2006), pp. G753-761 Conigrave, A.D.; Brown, E.M. & Rizzoli, R. (2008). Dietary protein and bone health: roles of

amino acid-sensing receptors in the control of calcium metabolism and bone

Ca2+-sensing receptor. *Proc Natl Acad Sci U S A*, Vol. 97, No. 9, (Apr 25 2000), pp.

J.M.; Tong, P.C.; Chen, Q.; Chan, V.A.; Reid, H.E.; Fazzalari, N.; Baker, H.M.; Baker, E.N.; Haggarty, N.W.; Grey, A.B. & Reid, I.R. (2004). Lactoferrin is a potent regulator of bone cell activity and increases bone formation in vivo. *Endocrinology*,

Strauss, B.J. (1998). Effects of dietary phytoestrogens in postmenopausal women.

Boutron-Ruault, M.C. (2008). Proteins, dietary acid load, and calcium and risk of postmenopausal fractures in the E3N French women prospective study. *J Bone* 


Arjmandi, B.H.; Birnbaum, R.; Goyal, N.V.; Getlinger, M.J.; Juma, S.; Alekel, L.; Hasler, C.M.;

Arjmandi, B.H.; Getlinger, M.J.; Goyal, N.V.; Alekel, L.; Hasler, C.M.; Juma, S.; Drum, M.L.;

Arjmandi, B.H.; Lucas, E.A.; Khalil, D.A.; Devareddy, L.; Smith, B.J.; McDonald, J.; Arquitt,

Arnett, T.R. (2008). Extracellular pH regulates bone cell function. *J Nutr*, Vol. 138, No. 2, (Feb

Ballard, T.L.; Clapper, J.A.; Specker, B.L.; Binkley, T.L. & Vukovich, M.D. (2005). Effect of

Barzel, U.S. & Massey, L.K. (1998). Excess dietary protein can adversely affect bone. *J Nutr*,

Beasley, J.M.; Ichikawa, L.E.; Ange, B.A.; Spangler, L.; LaCroix, A.Z.; Ott, S.M. & Scholes, D.

Bharadwaj, S.; Naidu, A.G.; Betageri, G.V.; Prasadarao, N.V. & Naidu, A.S. (2009). Milk

Bonjour, J.P. (2005). Dietary protein: an essential nutrient for bone health. *J Am Coll Nutr*,

Bonjour, J.P.; Ammann, P.; Chevalley, T. & Rizzoli, R. (2001). Protein intake and bone

Bos, K.J.; Rucklidge, G.J.; Dunbar, B. & Robins, S.P. (1999). Primary structure of the helical domain of porcine collagen X. *Matrix Biol*, Vol. 18, No. 2, (Apr 1999), pp. 149-153 Bouxsein, M.L. (2011). Bone structure and fracture risk: do they go arm in arm? *J Bone Miner* 

Bouxsein, M.L. & Seeman, E. (2009). Quantifying the material and structural determinants of bone strength. *Best Pract Res Clin Rheumatol*, Vol. 23, No. 6, (Dec 2009), pp. 741-753 Burr, D.B. (2002). The contribution of the organic matrix to bone's material properties. *Bone*,

Canaff, L.; Petit, J.L.; Kisiel, M.; Watson, P.H.; Gascon-Barre, M. & Hendy, G.N. (2001).

Extracellular calcium-sensing receptor is expressed in rat hepatocytes. coupling to

growth. *Can J Appl Physiol*, Vol. 26 Suppl, No. 2001), pp. S153-166

rats. *Am J Clin Nutr*, Vol. 68, No. 6 Suppl, (Dec 1998b), pp. 1358S-1363S Arjmandi, B.H.; Khalil, D.A.; Smith, B.J.; Lucas, E.A.; Juma, S.; Payton, M.E. & Wild, R.A.

*Nutr*, Vol. 68, No. 6 Suppl, (Dec 1998a), pp. 1364S-1368S

women. *Nutr J*, Vol. 4, No. 2005), pp. 8

Vol. 128, No. 6, (Jun 1998), pp. 1051-1053

*Clin Nutr*, Vol. 81, No. 6, (Jun 2005), pp. 1442-1448

*Metab*, Vol. 296, No. 6, (Jun 2009), pp. E1281-1288

Vol. 24, No. 6 Suppl, (Dec 2005), pp. 526S-536S

*Res*, Vol. 26, No. 7, (Jul 2011), pp. 1389-1391

Vol. 31, No. 1, (Jul 2002), pp. 8-11

*Am J Clin Nutr*, Vol. 91, No. 5, (May 2010), pp. 1311-1316

1048-1054

2008), pp. 415S-418S

Drum, M.L.; Hollis, B.W. & Kukreja, S.C. (1998a). Bone-sparing effect of soy protein in ovarian hormone-deficient rats is related to its isoflavone content. *Am J Clin* 

Hollis, B.W. & Kukreja, S.C. (1998b). Role of soy protein with normal or reduced isoflavone content in reversing bone loss induced by ovarian hormone deficiency in

(2003). Soy protein has a greater effect on bone in postmenopausal women not on hormone replacement therapy, as evidenced by reducing bone resorption and urinary calcium excretion. *J Clin Endocrinol Metab*, Vol. 88, No. 3, (Mar 2003), pp.

A.B.; Payton, M.E. & Mason, C. (2005). One year soy protein supplementation has positive effects on bone formation markers but not bone density in postmenopausal

protein supplementation during a 6-mo strength and conditioning program on insulin-like growth factor I and markers of bone turnover in young adults. *Am J* 

(2010). Is protein intake associated with bone mineral density in young women?

ribonuclease-enriched lactoferrin induces positive effects on bone turnover markers in postmenopausal women. *Osteoporos Int*, Vol. 20, No. 9, (Sep 2009), pp. 1603-1611 Blais, A.; Malet, A.; Mikogami, T.; Martin-Rouas, C. & Tome, D. (2009). Oral bovine

lactoferrin improves bone status of ovariectomized mice. *Am J Physiol Endocrinol* 

intracellular calcium mobilization and stimulation of bile flow. *J Biol Chem*, Vol. 276, No. 6, (Feb 9 2001), pp. 4070-4079


Dietary Protein and Bone Health 139

Fujimoto, K.; Tanaka, M.; Hirao, Y.; Nagata, Y.; Mori, M.; Miyanaga, N.; Akaza, H. & Kim,

Gahr, M.; Speer, C.P.; Damerau, B. & Sawatzki, G. (1991). Influence of lactoferrin on the

Gallagher, J.C.; Satpathy, R.; Rafferty, K. & Haynatzka, V. (2004). The effect of soy protein isolate on bone metabolism. *Menopause*, Vol. 11, No. 3, (May-Jun 2004), pp. 290-298 Geusens, P.P. & Boonen, S. (2002). Osteoporosis and the growth hormone-insulin-like

Giannini, S.; Nobile, M.; Dalle Carbonare, L.; Lodetti, M.G.; Sella, S.; Vittadello, G.; Minicuci,

Grassi, F.; Tell, G.; Robbie-Ryan, M.; Gao, Y.; Terauchi, M.; Yang, X.; Romanello, M.; Jones,

*Proc Natl Acad Sci U S A*, Vol. 104, No. 38, (Sep 18 2007), pp. 15087-15092 Grey, A.; Banovic, T.; Zhu, Q.; Watson, M.; Callon, K.; Palmano, K.; Ross, J.; Naot, D.; Reid,

Grey, A.; Zhu, Q.; Watson, M.; Callon, K. & Cornish, J. (2006). Lactoferrin potently inhibits

Guillerminet, F.; Beaupied, H.; Fabien-Soule, V.; Tome, D.; Benhamou, C.L.; Roux, C. &

Guillerminet, F.; Fabien-Soule, V.; Even, P.C.; Tome, D.; Benhamou, C.L.; Roux, C. & Blais,

Guo, H.Y.; Jiang, L.; Ibrahim, S.A.; Zhang, L.; Zhang, H.; Zhang, M. & Ren, F.Z. (2009).

He, F.J.; Marciniak, M.; Carney, C.; Markandu, N.D.; Anand, V.; Fraser, W.D.; Dalton, R.N.;

in mild hypertensives. *Hypertension*, Vol. 55, No. 3, (Mar 2010), pp. 681-688 Heaney, R.P. (2000). Dietary protein and phosphorus do not affect calcium absorption. *Am J* 

Heaney, R.P. (2007). Vitamin D endocrine physiology. *J Bone Miner Res*, Vol. 22 Suppl 2, No.

ovariectomized rats. *J Nutr*, Vol. 139, No. 5, (May 2009), pp. 958-964 Hannan, M.T.; Tucker, K.L.; Dawson-Hughes, B.; Cupples, L.A.; Felson, D.T. & Kiel, D.P.

growth factor axis. *Horm Res*, Vol. 58 Suppl 3, No. 2002), pp. 49-55

11, No. 3, 2008), pp. 252-257

2003), pp. 209-213

(Sep 2004), pp. 2268-2278

2504-2512

(Dec 2007), pp. V25-27

251, No. 1-2, (Jun 7 2006), pp. 96-102

*Bone*, Vol. 46, No. 3, (Mar 2010), pp. 827-834

*Clin Nutr*, Vol. 72, No. 3, (Sep 2000), pp. 758-761

ovariectomized C3H/HeN mice. *Osteoporos Int*, In Press

Vol. 49, No. 5, (May 1991), pp. 427-433

W.J. (2008). Age-stratified serum levels of isoflavones and proportion of equol producers in Japanese and Korean healthy men. *Prostate Cancer Prostatic Dis*, Vol.

function of human polymorphonuclear leukocytes and monocytes. *J Leukoc Biol*,

N. & Crepaldi, G. (2003). Hypercalciuria is a common and important finding in postmenopausal women with osteoporosis. *Eur J Endocrinol*, Vol. 149, No. 3, (Sep

D.P.; Weitzmann, M.N. & Pacifici, R. (2007). Oxidative stress causes bone loss in estrogen-deficient mice through enhanced bone marrow dendritic cell activation.

I.R. & Cornish, J. (2004). The low-density lipoprotein receptor-related protein 1 is a mitogenic receptor for lactoferrin in osteoblastic cells. *Mol Endocrinol*, Vol. 18, No. 9,

osteoblast apoptosis, via an LRP1-independent pathway. *Mol Cell Endocrinol*, Vol.

Blais, A. (2010). Hydrolyzed collagen improves bone metabolism and biomechanical parameters in ovariectomized mice: an in vitro and in vivo study.

A. (2011). Hydrolyzed collagen improves bone status and prevents bone loss in

Orally administered lactoferrin preserves bone mass and microarchitecture in

(2000). Effect of dietary protein on bone loss in elderly men and women: the Framingham Osteoporosis Study. *J Bone Miner Res*, Vol. 15, No. 12, (Dec 2000), pp.

Kaski, J.C. & MacGregor, G.A. (2010). Effects of potassium chloride and potassium bicarbonate on endothelial function, cardiovascular risk factors, and bone turnover


Darling, A.L.; Millward, D.J.; Torgerson, D.J.; Hewitt, C.E. & Lanham-New, S.A. (2009).

Dawson-Hughes, B. & Harris, S.S. (2002). Calcium intake influences the association of

Dawson-Hughes, B.; Harris, S.S.; Palermo, N.J.; Castaneda-Sceppa, C.; Rasmussen, H.M. &

Dawson-Hughes, B.; Harris, S.S.; Rasmussen, H.; Song, L. & Dallal, G.E. (2004). Effect of

Debbabi, H.; Dubarry, M.; Rautureau, M. & Tome, D. (1998). Bovine lactoferrin induces both

Delmi, M.; Rapin, C.H.; Bengoa, J.M.; Delmas, P.D.; Vasey, H. & Bonjour, J.P. (1990). Dietary

Devine, A.; Dick, I.M.; Islam, A.F.; Dhaliwal, S.S. & Prince, R.L. (2005). Protein consumption

Domrongkitchaiporn, S.; Pongskul, C.; Sirikulchayanonta, V.; Stitchantrakul, W.; Leeprasert,

Fenton, T.R.; Tough, S.C.; Lyon, A.W.; Eliasziw, M. & Hanley, D.A. (2011). Causal assessment

Frassetto, L.; Morris, R.C., Jr. & Sebastian, A. (2005). Long-term persistence of the urine

Frassetto, L.A.; Todd, K.M.; Morris, R.C., Jr. & Sebastian, A. (1998). Estimation of net

protein contents. *Am J Clin Nutr*, Vol. 68, No. 3, (Sep 1998), pp. 576-583

Hill's epidemiologic criteria for causality. *Nutr J*, Vol. 10, No. 2011), pp. 41 Feskanich, D.; Willett, W.C.; Stampfer, M.J. & Colditz, G.A. (1996). Protein consumption and bone fractures in women. *Am J Epidemiol*, Vol. 143, No. 5, (Mar 1 1996), pp. 472-479 Folman, Y. & Pope, G.S. (1969). Effect of norethisterone acetate, dimethylstilboestrol, genistein

immature mice. *J Endocrinol*, Vol. 44, No. 2, (Jun 1969), pp. 213-218

*Endocrinol Metab*, Vol. 90, No. 2, (Feb 2005), pp. 831-834

tubular acidosis. *Kidney Int*, Vol. 62, No. 6, (Dec 2002), pp. 2160-2166 Fenton, T.R.; Eliasziw, M.; Lyon, A.W.; Tough, S.C. & Hanley, D.A. (2008). Meta-analysis of

women. *J Clin Endocrinol Metab*, Vol. 89, No. 3, (Mar 2004), pp. 1169-1173 Dawson-Hughes, B.; Harris, S.S.; Rasmussen, H.M. & Dallal, G.E. (2007). Comparative

in humans. *Osteoporos Int*, Vol. 18, No. 7, (Jul 2007), pp. 955-961

*Nutr*, Vol. 90, No. 6, (Dec 2009), pp. 1674-1692

Vol. 75, No. 4, (Apr 2002), pp. 773-779

335, No. 8696, (Apr 28 1990), pp. 1013-1016

*Nutr*, Vol. 81, No. 6, (Jun 2005), pp. 1423-1428

1, (Jan 2009), pp. 96-102

1998), pp. 283-293

2008), pp. 1159-1166

Dietary protein and bone health: a systematic review and meta-analysis. *Am J Clin* 

protein intake with rates of bone loss in elderly men and women. *Am J Clin Nutr*,

Dallal, G.E. (2009). Treatment with potassium bicarbonate lowers calcium excretion and bone resorption in older men and women. *J Clin Endocrinol Metab*, Vol. 94, No.

dietary protein supplements on calcium excretion in healthy older men and

effects of oral aromatic and branched-chain amino acids on urine calcium excretion

mucosal and systemic immune response in mice. *J Dairy Res*, Vol. 65, No. 2, (May

supplementation in elderly patients with fractured neck of the femur. *Lancet*, Vol.

is an important predictor of lower limb bone mass in elderly women. *Am J Clin* 

V.; Ongphiphadhanakul, B.; Radinahamed, P. & Rajatanavin, R. (2002). Bone histology and bone mineral density after correction of acidosis in distal renal

the quantity of calcium excretion associated with the net acid excretion of the modern diet under the acid-ash diet hypothesis. *Am J Clin Nutr*, Vol. 88, No. 4, (Oct

of dietary acid load and bone disease: a systematic review & meta-analysis applying

and coumestrol on uptake of [3H]oestradiol by uterus, vagina and skeletal muscle of

calcium-lowering effect of potassium bicarbonate in postmenopausal women. *J Clin* 

endogenous noncarbonic acid production in humans from diet potassium and


Dietary Protein and Bone Health 141

Kerstetter, J.E.; O'Brien, K.O. & Insogna, K.L. (2003). Dietary protein, calcium metabolism,

Kerstetter, J.E.; Wall, D.E.; O'Brien, K.O.; Caseria, D.M. & Insogna, K.L. (2006). Meat and soy

Khalil, D.A.; Lucas, E.A.; Juma, S.; Smith, B.J.; Payton, M.E. & Arjmandi, B.H. (2002). Soy

Koh, W.P.; Wu, A.H.; Wang, R.; Ang, L.W.; Heng, D.; Yuan, J.M. & Yu, M.C. (2009). Gender-

Kreijkamp-Kaspers, S.; Kok, L.; Grobbee, D.E.; de Haan, E.H.; Aleman, A.; Lampe, J.W. &

Lean, J.M.; Davies, J.T.; Fuller, K.; Jagger, C.J.; Kirstein, B.; Partington, G.A.; Urry, Z.L. &

Legrand, D.; Elass, E.; Carpentier, M. & Mazurier, J. (2005). Lactoferrin: a modulator of

Legrand, D.; Elass, E.; Carpentier, M. & Mazurier, J. (2006). Interactions of lactoferrin with cells

Legrand, D.; Elass, E.; Pierce, A. & Mazurier, J. (2004). Lactoferrin and host defence: an

Lemann, J., Jr.; Bushinsky, D.A. & Hamm, L.L. (2003). Bone buffering of acid and base in humans. *Am J Physiol Renal Physiol*, Vol. 285, No. 5, (Nov 2003), pp. F811-832 Lemann, J., Jr.; Pleuss, J.A.; Gray, R.W. & Hoffmann, R.G. (1991). Potassium administration

adults [corrected]. *Kidney Int*, Vol. 39, No. 5, (May 1991), pp. 973-983 Lorget, F.; Clough, J.; Oliveira, M.; Daury, M.C.; Sabokbar, A. & Offord, E. (2002).

trial. *Eur J Nutr*, Vol. 43, No. 4, (Aug 2004), pp. 246-257

*Biochem Biophys Res Commun*, Vol. 296, No. 2, (Aug 16 2002), pp. 261-266 Lydeking-Olsen, E.; Beck-Jensen, J.E.; Setchell, K.D. & Holm-Jensen, T. (2004). Soymilk or

Lynch, M.P.; Stein, J.L.; Stein, G.S. & Lian, J.B. (1995). The influence of type I collagen on the

bone loss. *J Clin Invest*, Vol. 112, No. 6, (Sep 2003), pp. 915-923

Health Study. *Am J Epidemiol*, Vol. 170, No. 7, (Oct 1 2009), pp. 901-909 Koyama, Y.; Hirota, A.; Mori, H.; Takahara, H.; Kuwaba, K.; Kusubata, M.; Matsubara, Y.;

*Vitaminol (Tokyo)*, Vol. 47, No. 1, (Feb 2001), pp. 84-86

pp. 584S-592S

2006), pp. 1890-1895

1998), pp. 335-339

pp. 2549-2559

Vol. 17, No. 3, (Jun 2004), pp. 225-229

(Sep 2002), pp. 2605-2608

and skeletal homeostasis revisited. *Am J Clin Nutr*, Vol. 78, No. 3 Suppl, (Sep 2003),

protein affect calcium homeostasis in healthy women. *J Nutr*, Vol. 136, No. 7, (Jul

protein supplementation increases serum insulin-like growth factor-I in young and old men but does not affect markers of bone metabolism. *J Nutr*, Vol. 132, No. 9,

specific associations between soy and risk of hip fracture in the Singapore Chinese

Kasugai, S.; Itoh, M. & Irie, S. (2001). Ingestion of gelatin has differential effect on bone mineral density and body weight in protein undernutrition. *J Nutr Sci* 

van der Schouw, Y.T. (2004). Effect of soy protein containing isoflavones on cognitive function, bone mineral density, and plasma lipids in postmenopausal women: a randomized controlled trial. *Jama*, Vol. 292, No. 1, (Jul 7 2004), pp. 65-74 Lampe, J.W.; Karr, S.C.; Hutchins, A.M. & Slavin, J.L. (1998). Urinary equol excretion with a

soy challenge: influence of habitual diet. *Proc Soc Exp Biol Med*, Vol. 217, No. 3, (Mar

Chambers, T.J. (2003). A crucial role for thiol antioxidants in estrogen-deficiency

immune and inflammatory responses. *Cell Mol Life Sci*, Vol. 62, No. 22, (Nov 2005),

involved in immune function. *Biochem Cell Biol*, Vol. 84, No. 3, (Jun 2006), pp. 282-290

overview of its immuno-modulating and anti-inflammatory properties. *Biometals*,

reduces and potassium deprivation increases urinary calcium excretion in healthy

Lactoferrin reduces in vitro osteoclast differentiation and resorbing activity.

progesterone for prevention of bone loss--a 2 year randomized, placebo-controlled

development and maintenance of the osteoblast phenotype in primary and


Ho, S.C.; Woo, J.; Lam, S.; Chen, Y.; Sham, A. & Lau, J. (2003). Soy protein consumption and

Hunt, J.R.; Johnson, L.K. & Fariba Roughead, Z.K. (2009). Dietary protein and calcium

Ikeda, Y.; Iki, M.; Morita, A.; Kajita, E.; Kagamimori, S.; Kagawa, Y. & Yoneshima, H. (2006).

Ilich, J.Z.; Brownbill, R.A. & Tamborini, L. (2003). Bone and nutrition in elderly women:

Ince, B.A.; Anderson, E.J. & Neer, R.M. (2004). Lowering dietary protein to U.S. Recommended

Jackson, R.L.; Greiwe, J.S. & Schwen, R.J. (2011). Emerging evidence of the health benefits of

Jajoo, R.; Song, L.; Rasmussen, H.; Harris, S.S. & Dawson-Hughes, B. (2006). Dietary acid-

Kaneki, M.; Hodges, S.J.; Hosoi, T.; Fujiwara, S.; Lyons, A.; Crean, S.J.; Ishida, N.; Nakagawa,

Kerstetter, J.E. & Allen, L.H. (1990). Dietary protein increases urinary calcium. *J Nutr*, Vol.

Kerstetter, J.E.; Caseria, D.M.; Mitnick, M.E.; Ellison, A.F.; Gay, L.F.; Liskov, T.A.; Carpenter,

Kerstetter, J.E.; Mitnick, M.E.; Gundberg, C.M.; Caseria, D.M.; Ellison, A.F.; Carpenter, T.O.

Kerstetter, J.E.; O'Brien, K.O.; Caseria, D.M.; Wall, D.E. & Insogna, K.L. (2005). The impact of

Kerstetter, J.E.; O'Brien, K.O. & Insogna, K.L. (1998). Dietary protein affects intestinal calcium absorption. *Am J Clin Nutr*, Vol. 68, No. 4, (Oct 1998), pp. 859-865

women. *J Clin Endocrinol Metab*, Vol. 90, No. 1, (Jan 2005), pp. 26-31

(Oct 2003), pp. 835-842

432-448

(Jun 2006), pp. 224-230

(Apr 2001), pp. 315-321

pp. 1052-1055

120, No. 1, (Jan 1990), pp. 134-136

*Nutr*, Vol. 66, No. 5, (Nov 1997), pp. 1188-1196

Vol. 89, No. 5, (May 2009), pp. 1357-1365

*Nutr*, Vol. 136, No. 5, (May 2006), pp. 1323-1328

*Clin Nutr*, Vol. 57, No. 4, (Apr 2003), pp. 554-565

*Chem*, Vol. 53, No. 16, (Aug 10 2005), pp. 6531-6536

bone mass in early postmenopausal Chinese women. *Osteoporos Int*, Vol. 14, No. 10,

interact to influence calcium retention: a controlled feeding study. *Am J Clin Nutr*,

Intake of fermented soybeans, natto, is associated with reduced bone loss in postmenopausal women: Japanese Population-Based Osteoporosis (JPOS) Study. *J* 

protein, energy, and calcium as main determinants of bone mineral density. *Eur J* 

dietary allowance levels reduces urinary calcium excretion and bone resorption in young women. *J Clin Endocrinol Metab*, Vol. 89, No. 8, (Aug 2004), pp. 3801-3807 Iwai, K.; Hasegawa, T.; Taguchi, Y.; Morimatsu, F.; Sato, K.; Nakamura, Y.; Higashi, A.;

Kido, Y.; Nakabo, Y. & Ohtsuki, K. (2005). Identification of food-derived collagen peptides in human blood after oral ingestion of gelatin hydrolysates. *J Agric Food* 

S-equol, an estrogen receptor beta agonist. *Nutr Rev*, Vol. 69, No. 8, (Aug 2011), pp.

base balance, bone resorption, and calcium excretion. *J Am Coll Nutr*, Vol. 25, No. 3,

M.; Takechi, M.; Sano, Y.; Mizuno, Y.; Hoshino, S.; Miyao, M.; Inoue, S.; Horiki, K.; Shiraki, M.; Ouchi, Y. & Orimo, H. (2001). Japanese fermented soybean food as the major determinant of the large geographic difference in circulating levels of vitamin K2: possible implications for hip-fracture risk. *Nutrition*, Vol. 17, No. 4,

T.O. & Insogna, K.L. (1997). Increased circulating concentrations of parathyroid hormone in healthy, young women consuming a protein-restricted diet. *Am J Clin* 

& Insogna, K.L. (1999). Changes in bone turnover in young women consuming different levels of dietary protein. *J Clin Endocrinol Metab*, Vol. 84, No. 3, (Mar 1999),

dietary protein on calcium absorption and kinetic measures of bone turnover in


Dietary Protein and Bone Health 143

Munger, R.G.; Cerhan, J.R. & Chiu, B.C. (1999). Prospective study of dietary protein intake

Muthusami, S.; Ramachandran, I.; Muthusamy, B.; Vasudevan, G.; Prabhu, V.;

Naot, D.; Grey, A.; Reid, I.R. & Cornish, J. (2005). Lactoferrin--a novel bone growth factor.

New, S.A.; MacDonald, H.M.; Campbell, M.K.; Martin, J.C.; Garton, M.J.; Robins, S.P. &

Nomura, Y.; Oohashi, K.; Watanabe, M. & Kasugai, S. (2005). Increase in bone mineral

Oesser, S.; Adam, M.; Babel, W. & Seifert, J. (1999). Oral administration of (14)C labeled

Oh, S.M.; Lee, S.H.; Lee, B.J.; Pyo, C.W.; Yoo, N.K.; Lee, S.Y.; Kim, J. & Choi, S.Y. (2007). A

Ohara, H.; Matsumoto, H.; Ito, K.; Iwai, K. & Sato, K. (2007). Comparison of quantity and

Ohlsson, C.; Mohan, S.; Sjogren, K.; Tivesten, A.; Isgaard, J.; Isaksson, O.; Jansson, J.O. &

Oreffo, R.O.; Mundy, G.R.; Seyedin, S.M. & Bonewald, L.F. (1989). Activation of the bone-

Owen, T.A.; Aronow, M.S.; Barone, L.M.; Bettencourt, B.; Stein, G.S. & Lian, J.B. (1991).

Perrini, S.; Laviola, L.; Carreira, M.C.; Cignarelli, A.; Natalicchio, A. & Giorgino, F. (2010).

(Jan 1999), pp. 147-152

Vol. 360, No. 1-2, (Oct 2005), pp. 81-86

*Clin Med Res*, Vol. 3, No. 2, (May 2005), pp. 93-101

*Nephrol*, Vol. 17, No. 3, (Mar 2006), pp. 617-626

*Nutrition*, Vol. 21, No. 11-12, (Nov-Dec 2005), pp. 1120-1126

(C57/BL). *J Nutr*, Vol. 129, No. 10, (Oct 1999), pp. 1891-1895

*Immunol*, Vol. 179, No. 9, (Nov 1 2007), pp. 5686-5692

No. 4, (Feb 21 2007), pp. 1532-1535

205, No. 3, (Jun 2010), pp. 201-210

1496-1504

*Rev*, Vol. 30, No. 5, (Aug 2009), pp. 494-535

*Commun*, Vol. 158, No. 3, (Feb 15 1989), pp. 817-823

and risk of hip fracture in postmenopausal women. *Am J Clin Nutr*, Vol. 69, No. 1,

Subramaniam, V.; Jagadeesan, A. & Narasimhan, S. (2005). Ovariectomy induces oxidative stress and impairs bone antioxidant system in adult rats. *Clin Chim Acta*,

Reid, D.M. (2004). Lower estimates of net endogenous non-carbonic acid production are positively associated with indexes of bone health in premenopausal and perimenopausal women. *Am J Clin Nutr*, Vol. 79, No. 1, (Jan 2004), pp. 131-138 Nijenhuis, T.; Renkema, K.Y.; Hoenderop, J.G. & Bindels, R.J. (2006). Acid-base status

determines the renal expression of Ca2+ and Mg2+ transport proteins. *J Am Soc* 

density through oral administration of shark gelatin to ovariectomized rats.

gelatin hydrolysate leads to an accumulation of radioactivity in cartilage of mice

distinct role of neutrophil lactoferrin in RelA/p65 phosphorylation on Ser536 by recruiting TNF receptor-associated factors to IkappaB kinase signaling complex. *J* 

structures of hydroxyproline-containing peptides in human blood after oral ingestion of gelatin hydrolysates from different sources. *J Agric Food Chem*, Vol. 55,

Svensson, J. (2009). The role of liver-derived insulin-like growth factor-I. *Endocr* 

derived latent TGF beta complex by isolated osteoclasts. *Biochem Biophys Res* 

Pleiotropic effects of vitamin D on osteoblast gene expression are related to the proliferative and differentiated state of the bone cell phenotype: dependency upon basal levels of gene expression, duration of exposure, and bone matrix competency in normal rat osteoblast cultures. *Endocrinology*, Vol. 128, No. 3, (Mar 1991), pp.

The GH/IGF1 axis and signaling pathways in the muscle and bone: mechanisms underlying age-related skeletal muscle wasting and osteoporosis. *J Endocrinol*, Vol.

passaged rat calvarial osteoblasts: modification of expression of genes supporting cell growth, adhesion, and extracellular matrix mineralization. *Exp Cell Res*, Vol. 216, No. 1, (Jan 1995), pp. 35-45


Malet, A.; Bournaud, E.; Lan, A.; Mikogami, T.; Tome, D. & Blais, A. (2011). Bovine

Mallegol, J.; van Niel, G. & Heyman, M. (2005). Phenotypic and functional characterization

Mann, V.; Hobson, E.E.; Li, B.; Stewart, T.L.; Grant, S.F.; Robins, S.P.; Aspden, R.M. &

Marie, P.J. (2010). The calcium-sensing receptor in bone cells: a potential therapeutic target

Marini, H.; Minutoli, L.; Polito, F.; Bitto, A.; Altavilla, D.; Atteritano, M.; Gaudio, A.;

Massey, L.K. (2003). Dietary animal and plant protein and human bone health: a whole

Massey, L.K. & Kynast-Gales, S.A. (2001). Diets with either beef or plant proteins reduce risk

Maurer, M.; Riesen, W.; Muser, J.; Hulter, H.N. & Krapf, R. (2003). Neutralization of Western

in humans. *Am J Physiol Renal Physiol*, Vol. 284, No. 1, (Jan 2003), pp. F32-40 Mazess, R.B. & Barden, H.S. (1991). Bone density in premenopausal women: effects of age,

Messina, M. (2010). Insights gained from 20 years of soy research. *J Nutr*, Vol. 140, No. 12,

Metz, J.A.; Anderson, J.J. & Gallagher, P.N., Jr. (1993). Intakes of calcium, phosphorus, and

Meyer, H.E.; Pedersen, J.I.; Loken, E.B. & Tverdal, A. (1997). Dietary factors and the

Morita, Y.; Matsuyama, H.; Serizawa, A.; Takeya, T. & Kawakami, H. (2008). Identification

Morton, M.S.; Arisaka, O.; Miyake, N.; Morgan, L.D. & Evans, B.A. (2002). Phytoestrogen

women. *Am J Clin Nutr*, Vol. 58, No. 4, (Oct 1993), pp. 537-542

*Epidemiol*, Vol. 145, No. 2, (Jan 15 1997), pp. 117-123

*Bone*, Vol. 42, No. 2, (Feb 2008), pp. 380-387

*Nutr*, Vol. 132, No. 10, (Oct 2002), pp. 3168-3171

foods approach. *J Nutr*, Vol. 133, No. 3, (Mar 2003), pp. 862S-865S

*J Am Diet Assoc*, Vol. 101, No. 3, (Mar 2001), pp. 326-331

Vol. 53, No. 1, (Jan 1991), pp. 132-142

(Dec 2010), pp. 2289S-2295S

modulation. *Bone*, Vol. 48, No. 5, (May 1 2011), pp. 1028-1035

in osteoporosis. *Bone*, Vol. 46, No. 3, (Mar 2010), pp. 571-576

216, No. 1, (Jan 1995), pp. 35-45

No. 7, (Apr 2001), pp. 899-907

pp. 11-16

847

passaged rat calvarial osteoblasts: modification of expression of genes supporting cell growth, adhesion, and extracellular matrix mineralization. *Exp Cell Res*, Vol.

lactoferrin improves bone status of ovariectomized mice via immune function

of intestinal epithelial exosomes. *Blood Cells Mol Dis*, Vol. 35, No. 1, (Jul-Aug 2005),

Ralston, S.H. (2001). A COL1A1 Sp1 binding site polymorphism predisposes to osteoporotic fracture by affecting bone density and quality. *J Clin Invest*, Vol. 107,

Mazzaferro, S.; Frisina, A.; Frisina, N.; Lubrano, C.; Bonaiuto, M.; D'Anna, R.; Cannata, M.L.; Corrado, F.; Adamo, E.B.; Wilson, S. & Squadrito, F. (2007). Effects of the phytoestrogen genistein on bone metabolism in osteopenic postmenopausal women: a randomized trial. *Ann Intern Med*, Vol. 146, No. 12, (Jun 19 2007), pp. 839-

of calcium oxalate precipitation in patients with a history of calcium kidney stones.

diet inhibits bone resorption independently of K intake and reduces cortisol secretion

dietary intake, physical activity, smoking, and birth-control pills. *Am J Clin Nutr*,

protein, and physical-activity level are related to radial bone mass in young adult

incidence of hip fracture in middle-aged Norwegians. A prospective study. *Am J* 

of angiogenin as the osteoclastic bone resorption-inhibitory factor in bovine milk.

concentrations in serum from Japanese men and women over forty years of age. *J* 


Dietary Protein and Bone Health 145

Shanker, G.; Sorci-Thomas, M. & Adams, M.R. (1994). Estrogen modulates the expression of

THP-1 cells. *Lymphokine Cytokine Res*, Vol. 13, No. 6, (Dec 1994), pp. 377-382 Stein, G.S. & Lian, J.B. (1993). Molecular mechanisms mediating proliferation/differentiation

Suda, T.; Takahashi, N.; Udagawa, N.; Jimi, E.; Gillespie, M.T. & Martin, T.J. (1999).

Takayama, Y. & Mizumachi, K. (2008). Effect of bovine lactoferrin on extracellular matrix

Takeuchi, Y.; Nakayama, K. & Matsumoto, T. (1996). Differentiation and cell surface

Takeuchi, Y.; Suzawa, M.; Kikuchi, T.; Nishida, E.; Fujita, T. & Matsumoto, T. (1997).

Teegarden, D.; Lyle, R.M.; McCabe, G.P.; McCabe, L.D.; Proulx, W.R.; Michon, K.; Knight,

Thorpe, D.L.; Knutsen, S.F.; Beeson, W.L.; Rajaram, S. & Fraser, G.E. (2008a). Effects of meat

Thorpe, M. & Evans, E.M. (2011). Dietary protein and bone health: harmonizing conflicting

Thorpe, M.; Mojtahedi, M.C.; Chapman-Novakofski, K.; McAuley, E. & Evans, E.M. (2008b).

Tkatch, L.; Rapin, C.H.; Rizzoli, R.; Slosman, D.; Nydegger, V.; Vasey, H. & Bonjour, J.P.

the proximal femur. *J Am Coll Nutr*, Vol. 11, No. 5, (Oct 1992), pp. 519-525 Turner, C.H. (2006). Bone strength: current concepts. *Ann N Y Acad Sci*, Vol. 1068, No. (Apr

Uenishi, K.; Ishida, H.; Toba, Y.; Aoe, S.; Itabashi, A. & Takada, Y. (2007). Milk basic protein

Vatanparast, H.; Bailey, D.A.; Baxter-Jones, A.D. & Whiting, S.J. (2007). The effects of dietary

women. *Osteoporos Int*, Vol. 18, No. 3, (Mar 2007), pp. 385-390

calcium intake. *J Nutr*, Vol. 137, No. 12, (Dec 2007), pp. 2674-2679

*Endocr Rev*, Vol. 14, No. 4, (Aug 1993), pp. 424-442

1999), pp. 345-357

(Jan 2008), pp. 226-230

No. 7, (Feb 16 1996), pp. 3938-3944

(Nov 14 1997), pp. 29309-29316

pp. 564-572

(Jan 2008b), pp. 80-85

2006), pp. 429-446

*Clin Nutr*, Vol. 68, No. 3, (Sep 1998), pp. 749-754

theories. *Nutr Rev*, Vol. 69, No. 4, (Apr 2011), pp. 215-230

tumor necrosis factor alpha mRNA in phorbol ester-stimulated human monocytic

interrelationships during progressive development of the osteoblast phenotype.

Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. *Endocr Rev*, Vol. 20, No. 3, (Jun

calcification by human osteoblast-like cells. *Biosci Biotechnol Biochem*, Vol. 72, No. 1,

expression of transforming growth factor-beta receptors are regulated by interaction with matrix collagen in murine osteoblastic cells. *J Biol Chem*, Vol. 271,

Differentiation and transforming growth factor-beta receptor down-regulation by collagen-alpha2beta1 integrin interaction is mediated by focal adhesion kinase and its downstream signals in murine osteoblastic cells. *J Biol Chem*, Vol. 272, No. 46,

A.P.; Johnston, C.C. & Weaver, C.M. (1998). Dietary calcium, protein, and phosphorus are related to bone mineral density and content in young women. *Am J* 

consumption and vegetarian diet on risk of wrist fracture over 25 years in a cohort of peri- and postmenopausal women. *Public Health Nutr*, Vol. 11, No. 6, (Jun 2008a),

A positive association of lumbar spine bone mineral density with dietary protein is suppressed by a negative association with protein sulfur. *J Nutr*, Vol. 138, No. 1,

(1992). Benefits of oral protein supplementation in elderly patients with fracture of

increases bone mineral density and improves bone metabolism in healthy young

protein on bone mineral mass in young adults may be modulated by adolescent


Potter, S.M.; Baum, J.A.; Teng, H.; Stillman, R.J.; Shay, N.F. & Erdman, J.W., Jr. (1998). Soy

Promislow, J.H.; Goodman-Gruen, D.; Slymen, D.J. & Barrett-Connor, E. (2002). Protein

Quarles, L.D.; Yohay, D.A.; Lever, L.W.; Caton, R. & Wenstrup, R.J. (1992). Distinct

Rahbar, A.; Larijani, B.; Nabipour, I.; Mohamadi, M.M.; Mirzaee, K. & Amiri, Z. (2009).

Ramshaw, J.A.; Shah, N.K. & Brodsky, B. (1998). Gly-X-Y tripeptide frequencies in collagen:

Rapuri, P.B.; Gallagher, J.C. & Haynatzka, V. (2003). Protein intake: effects on bone mineral

Reinwald, S. & Weaver, C.M. (2010). Soy components vs. whole soy: are we betting our bones on a long shot? *J Nutr*, Vol. 140, No. 12, (Dec 2010), pp. 2312S-2317S Roughead, Z.K.; Johnson, L.K.; Lykken, G.I. & Hunt, J.R. (2003). Controlled high meat diets

Schurch, M.A.; Rizzoli, R.; Slosman, D.; Vadas, L.; Vergnaud, P. & Bonjour, J.P. (1998).

Schwartz, A.V.; Kelsey, J.L.; Maggi, S.; Tuttleman, M.; Ho, S.C.; Jonsson, P.V.; Poor, G.;

Seeman, E. & Delmas, P.D. (2006). Bone quality--the material and structural basis of bone strength and fragility. *N Engl J Med*, Vol. 354, No. 21, (May 25 2006), pp. 2250-2261 Sellmeyer, D.E.; Stone, K.L.; Sebastian, A. & Cummings, S.R. (2001). A high ratio of dietary

Setchell, K.D.; Brown, N.M. & Lydeking-Olsen, E. (2002). The clinical importance of the

*Am J Epidemiol*, Vol. 155, No. 7, (Apr 1 2002), pp. 636-644

population. *Bone*, Vol. 45, No. 5, (Nov 2009), pp. 876-881

women. *J Nutr*, Vol. 133, No. 4, (Apr 2003), pp. 1020-1026

1375S-1379S

pp. 683-692

pp. 86-91

(Jun 2003), pp. 1517-1525

10, (May 15 1998), pp. 801-809

*Osteoporos Int*, Vol. 9, No. 3, 1999), pp. 242-253

*Clin Nutr*, Vol. 73, No. 1, (Jan 2001), pp. 118-122

132, No. 12, (Dec 2002), pp. 3577-3584

protein and isoflavones: their effects on blood lipids and bone density in postmenopausal women. *Am J Clin Nutr*, Vol. 68, No. 6 Suppl, (Dec 1998), pp.

consumption and bone mineral density in the elderly : the Rancho Bernardo Study.

proliferative and differentiated stages of murine MC3T3-E1 cells in culture: an in vitro model of osteoblast development. *J Bone Miner Res*, Vol. 7, No. 6, (Jun 1992),

Relationship among dietary estimates of net endogenous acid production, bone mineral density and biochemical markers of bone turnover in an Iranian general

a context for host-guest triple-helical peptides. *J Struct Biol*, Vol. 122, No. 1-2, 1998),

density and the rate of bone loss in elderly women. *Am J Clin Nutr*, Vol. 77, No. 6,

do not affect calcium retention or indices of bone status in healthy postmenopausal

Protein supplements increase serum insulin-like growth factor-I levels and attenuate proximal femur bone loss in patients with recent hip fracture. A randomized, double-blind, placebo-controlled trial. *Ann Intern Med*, Vol. 128, No.

Sisson de Castro, J.A.; Xu, L.; Matkin, C.C.; Nelson, L.M. & Heyse, S.P. (1999). International variation in the incidence of hip fractures: cross-national project on osteoporosis for the World Health Organization Program for Research on Aging.

animal to vegetable protein increases the rate of bone loss and the risk of fracture in postmenopausal women. Study of Osteoporotic Fractures Research Group. *Am J* 

metabolite equol-a clue to the effectiveness of soy and its isoflavones. *J Nutr*, Vol.


Weikert, C.; Walter, D.; Hoffmann, K.; Kroke, A.; Bergmann, M.M. & Boeing, H. (2005). The

Wengreen, H.J.; Munger, R.G.; West, N.A.; Cutler, D.R.; Corcoran, C.D.; Zhang, J. & Sassano,

residents of Utah. *J Bone Miner Res*, Vol. 19, No. 4, (Apr 2004), pp. 537-545 Whiting, S.J.; Anderson, D.J. & Weeks, S.J. (1997). Calciuric effects of protein and potassium

Wilson, J. & Wilson, G.J. (2006). Contemporary issues in protein requirements and

Wu, J.; Fujioka, M.; Sugimoto, K.; Mu, G. & Ishimi, Y. (2004). Assessment of effectiveness of

Wynn, E.; Lanham-New, S.A.; Krieg, M.A.; Whittamore, D.R. & Burckhardt, P. (2008). Low

Xiao, G.; Wang, D.; Benson, M.D.; Karsenty, G. & Franceschi, R.T. (1998). Role of the alpha2-

Zhang, X.; Shu, X.O.; Li, H.; Yang, G.; Li, Q.; Gao, Y.T. & Zheng, W. (2005). Prospective cohort

Zou, Z.Y.; Lin, X.M.; Xu, X.R.; Xu, R.; Ma, L.; Li, Y. & Wang, M.F. (2009). Evaluation of milk

Zwart, S.R.; Davis-Street, J.E.; Paddon-Jones, D.; Ferrando, A.A.; Wolfe, R.R. & Smith, S.M.

women. *Arch Intern Med*, Vol. 165, No. 16, (Sep 12 2005), pp. 1890-1895 Zhou, S.; Turgeman, G.; Harris, S.E.; Leitman, D.C.; Komm, B.S.; Bodine, P.V. & Gazit, D.

young women. *Eur J Nutr*, Vol. 48, No. 5, (Aug 2009), pp. 301-306

weightlessness. *J Appl Physiol*, Vol. 99, No. 1, (Jul 2005), pp. 134-140

rats. *J Bone Miner Metab*, Vol. 22, No. 6, 2004), pp. 547-553

No. (Nov 2007), pp. 360-375

pp. 402-409

pp. 7-27

(Jul 2008), pp. 1349-1354

relation between dietary protein, calcium and bone health in women: results from the EPIC-Potsdam cohort. *Ann Nutr Metab*, Vol. 49, No. 5, (Sep-Oct 2005), pp. 312-318 Weitzmann, M.N. & Pacifici, R. (2007). T cells: unexpected players in the bone loss induced

by estrogen deficiency and in basal bone homeostasis. *Ann N Y Acad Sci*, Vol. 1116,

N.E. (2004). Dietary protein intake and risk of osteoporotic hip fracture in elderly

bicarbonate but not of sodium chloride or phosphate can be detected acutely in adult women and men. *Am J Clin Nutr*, Vol. 65, No. 5, (May 1997), pp. 1465-1472 Whiting, S.J.; Boyle, J.L.; Thompson, A.; Mirwald, R.L. & Faulkner, R.A. (2002). Dietary

protein, phosphorus and potassium are beneficial to bone mineral density in adult men consuming adequate dietary calcium. *J Am Coll Nutr*, Vol. 21, No. 5, (Oct 2002),

consumption for resistance trained athletes. *J Int Soc Sports Nutr*, Vol. 3, No. 2006),

oral administration of collagen peptide on bone metabolism in growing and mature

estimates of dietary acid load are positively associated with bone ultrasound in women older than 75 years of age with a lifetime fracture. *J Nutr*, Vol. 138, No. 7,

integrin in osteoblast-specific gene expression and activation of the Osf2 transcription factor. *J Biol Chem*, Vol. 273, No. 49, (Dec 4 1998), pp. 32988-32994 Yamamura, J.; Aoe, S.; Toba, Y.; Motouri, M.; Kawakami, H.; Kumegawa, M.; Itabashi, A. &

Takada, Y. (2002). Milk basic protein (MBP) increases radial bone mineral density in healthy adult women. *Biosci Biotechnol Biochem*, Vol. 66, No. 3, (Mar 2002), pp. 702-704

study of soy food consumption and risk of bone fracture among postmenopausal

(2003). Estrogens activate bone morphogenetic protein-2 gene transcription in mouse mesenchymal stem cells. *Mol Endocrinol*, Vol. 17, No. 1, (Jan 2003), pp. 56-66

basic protein supplementation on bone density and bone metabolism in Chinese

(2005). Amino acid supplementation alters bone metabolism during simulated

### *Edited by Abdellah El Maghraoui*

The World Health Organization (WHO) has established dual-energy x-ray absorptiometry (DXA) as the best densitometric technique for assessing bone mineral density (BMD) in postmenopausal women and has based the definitions of osteopenia and osteoporosis on its results. DXA enables accurate diagnosis of osteoporosis, estimation of fracture risk and monitoring of patients undergoing treatment. Additional features of DXA include measurement of BMD at multiple skeletal sites, vertebral fracture assessment and body composition assessment, including fat mass and lean soft tissue mass of the whole body and the segments. This book contains reviews and original studies about DXA and its different uses in clinical practice (diagnosis of osteoporosis, monitoring of BMD measurement) and in medical research in several situations (e.g. assessment of morphological asymmetry in athletes, estimation of resting energy expenditure, assessment of vertebral strength and vertebral fracture risk, or study of dry bones such as the ulna).

ISBN 978-953-307-877-9

ISBN 978-953-51-6742-6

Dual Energy X-Ray Absorptiometry

Dual Energy X-Ray

Absorptiometry

*Edited by Abdellah El Maghraoui*

Photo by Rost-9D / iStock