Bone Size, Architecture and Strength Deficits in Cerebral Palsy
Childhood and adolescence are the most important periods of bone development. The bones experience an increase in size, an accretion of mineral, a change in architecture and an increase in strength that accommodate the greater strains on the skeleton due to the increasing weight of the growing body and the increasing muscle forces produced during physical activity. Children with cerebral palsy (CP) tend to have smaller bones, lower areal bone mineral density, and less developed bone architecture compared to typically developing children, which contributes to their much weaker bones and higher risk for fracture. In this chapter, we will review how deficits in bone that emerge during the growth and development of children with CP contribute to their high rate of fragility fractures. We will also review methods of bone assessment, factors that contribute to the poor bone development in children and adolescents with CP, bone health in adults with CP, and potential treatment strategies.
KeywordsCerebral palsy Bone structure Bone strength Fracture Unloading
During childhood and adolescence, there is a tremendous growth and development of the bones that comprise the skeleton (Heaney et al. 2000; Modlesky and Lewis 2002). It has been proposed that environmental factors have a considerable influence on the bone changes that occur during this early period of life (Parfitt 1994). For example, regular mechanical loading and adequate nutritional intake are necessary to facilitate the expected increases in mass and the changes in composition and architecture of bone during the first two decades of life (Heaney et al. 2000; Modlesky and Lewis 2002). Because of their issues with movement and posture, children with cerebral palsy (CP) do not have the same level of mechanical loading as typically developing children (Bjornson et al. 2007; Johnson et al. 2009; Modlesky et al. 2009). Furthermore, some children with CP do not receive the same nutritional intake (Henderson et al. 1995, 2002a). As a result, atypical bone growth and development are common (Binkley et al. 2005; Henderson et al. 2002a; Modlesky et al. 2008, 2009, 2015; Whitney et al. 2017) and the incidence of fragility, or low-energy, fractures is increased, especially in the lower extremity bones of children with CP who are nonambulatory (McIvor and Samilson 1966; Presedo et al. 2007). Poor bone growth and development during childhood likely carries through adulthood and increases the risk of fractures later in life (Heaney et al. 2000; Modlesky and Lewis 2002; Wren et al. 2014). In this chapter, we will review typical bone growth and development, bone growth and development in children with CP, bone assessment techniques, fractures in children with CP, factors that contribute to atypical bone growth and development in children with CP, bone health in adults with CP, and potential treatment strategies.
Bone Anatomy and Typical Bone Growth and Development
Bone modeling is the primary process that facilitates the growth and development of bone during infancy, childhood, and adolescence (Parfitt 1994). The activity of bone modeling is carried out by chondrocytes, osteoblasts, and osteoclasts. The chondrocytes in the epiphyseal (growth) plate develop and expand. The chondrocytes that are close to the diaphysis attract osteoblasts and lead to the production of new bone. The activities of the chondrocytes and osteoblasts facilitate the elongation of bone until the body matures and the epiphyseal plates fuse and stop producing new chondrocytes. During the modeling process, osteoclasts are also involved, but they are not necessarily coupled with the osteoblasts. For example, as osteoblasts are working with chondrocytes to elongate the bone and to add bone on the periosteal surface to increase the width of the bone diaphysis, there is osteoclast activity on the endosteal surface leading to an expansion of the medullary cavity (Marks and Odgren 2002). Overall, there is greater osteoblast activity than osteoclast activity during modeling, which leads to an increase in the thickness of the cortical walls and the thickness of the trabeculae (Kirmani et al. 2009), as well as the size of the bones and the overall skeleton. Peak bone mass is reached during the third decade of life (Heaney et al. 2000).
Bone remodeling, a secondary process during growth and development, plays a much larger role in the mature skeleton (Parfitt 1994). During bone remodeling, osteoclasts and osteoblasts work closely together to replace old bone tissue with new bone tissue (Langdahl et al. 2016). Osteocytes, although not directly involved in bone tissue formation or resorption, play an important role in translating mechanical loading into biochemical signaling and thus help regulate the modeling and remodeling processes (Bonewald and Johnson 2008). Throughout the growth period, bones change in size and shape to accommodate the mechanical stimulation created by gravitational forces and increased muscle action on bone.
Bone Growth and Development in Children with CP
There is a restriction in the growth and development of bone in children with CP. Bone length is stunted as reflected by the 16% shorter bones and 13% shorter height observed in children with CP who are unable to ambulate (i.e., Gross Motor Function Classification, GMFCS = IV-V) or are only able to ambulate with assistance (i.e., GMFCS = III) compared to children with typical development (Krick et al. 1996; Modlesky et al. 2009) and the >2 SD shorter height than normative values in infants with CP (Krick et al. 1996). In addition to shorter bones, children with CP have much lower areal bone mineral density (aBMD) and bone mineral content in the distal femur (Henderson et al. 2002a; King et al. 2003; Modlesky et al. 2008), as assessed by dual-energy X-ray absorptiometry (DXA). Low aBMD has also been reported in the lumbar spine, though the discrepancy is not as marked. In children with CP and a GMFCS III-V, aBMD z-scores of −3.1 in the distal femur and −1.8 in the lumbar spine have been reported (Henderson et al. 2002a). The compromised was the largest in children who had the most severe motor deficits (GMFCS V) and the smallest in the children with the least severe motor deficits (GMFCS III) (Henderson et al. 2002a).
There is evidence that there is no compromise in the material density of bone in children with CP and it may even be higher in children with CP than in typically developing children at greater cortical widths (Binkley et al. 2005). However, aBMD assessed by DXA is lower in children with CP than in children with typical development (Henderson et al. 2002a; Modlesky et al. 2008). The lower aBMD is a reflection of the smaller bones in children with CP. Interestingly, children with CP also have a higher concentration of fat in their bone marrow (Whitney et al. 2017), which may indicate a greater propensity of mesenchymal stem cells to form adipocytes rather than osteoblasts due to the limited mechanical loading associated with CP (Luu et al. 2009).
High Rate of Fragility Fractures in Children with CP
Due to the atypical growth and development of bone in children and adolescents with CP, fragility fractures are a significant complication. Fractures are most common in children with CP who have the most limited mobility; they occur in approximately one quarter of children with CP with a GMFCS III-V and >10 years of age (Henderson et al. 2002a). Repeat fractures are often experienced later in life (Henderson 1997; Lee and Lyne 1990; McIvor and Samilson 1966). Most fractures in children with CP occur in the lower extremities (McIvor and Samilson 1966, Presedo et al. 2007). The femur is the most commonly fractured bone, and the distal femur is the most common fracture site (McIvor and Samilson 1966, Presedo et al. 2007). While femur fractures represent only 2% of all fractures in children without physical disabilities (Worlock and Stower 1986), they represent half or more of all fractures in nonambulatory children with CP (Henderson 1997; McIvor and Samilson 1966; Presedo et al. 2007).
Assessing Bone in Children with CP
Areal bone mineral density assessed by DXA is the single best measure used to assess fracture risk that is widely available (Genant et al. 1996) and it has been shown to predict fracture in the distal femur of children and adolescents with CP and a GMFCS III-V (Henderson et al. 2010). On the other hand, aBMD is an imperfect surrogate of bone strength and fracture risk, especially in children. Moreover, there is a significant overlap in aBMD from DXA in those who do and do not fracture (Ciarelli et al. 2000; Kleerekoper et al. 1985; Majumdar et al. 1999; Ott 1993). Up to 50% of the variance in bone strength is explained by other features of the skeleton (Ciarelli et al. 2000; Dempster 2000; Link et al. 1998; Majumdar et al. 1999; Parfitt 1987). Furthermore, because it is affected by bone size, using aBMD alone to make clinical decisions in children has been questioned (Klein et al. 2005; Lewiecki et al. 2004). Bone mineral content controlled for body size is viewed as a more appropriate measure to assess fracture risk in children (Heaney 2004; Prentice et al. 1994). However, it also has limitations.
One key feature that should be considered when evaluating bone status in children (and adults) is bone architecture, which is also referred to as bone structure (Ciarelli et al. 2000; Dempster 2000; Kleerekoper et al. 1985; Link et al. 1998; Majumdar et al. 1999; Parfitt 1987). This notion is supported by strong evidence that bone architecture can discriminate between those who do and do not fracture (Ciarelli et al. 2000; Dempster 2000; Jamal et al. 2006; Kleerekoper et al. 1985; Link et al. 1998; Majumdar et al. 1999; Parfitt 1987; Sornay-Rendu et al. 2007). Furthermore, there is an improvement in the prediction of strength and fracture when measures of bone mass are combined with measures of trabecular bone microarchitecture, such as trabecular bone volume to total volume, trabecular number, trabecular thickness and trabecular separation (Majumdar et al. 1999; Siffert et al. 1996), or measures of cortical bone architecture (Augat and Schorlemmer 2006; Bousson et al. 2006; Laib et al. 2001; Sell et al. 2005).
Magnetic resonance imaging (MRI) and computed tomography provide accurate estimates of bone architecture in humans (Boutroy et al. 2005; Majumdar et al. 1998; Woodhead et al. 2001). Magnetic resonance imaging is particularly attractive when working with children with CP because there is no ionizing radiation. The main challenges associated with the widespread use of MRI include the high expense, the limited available technological expertise and the potential involuntary movement of some individuals with CP due to spasticity and behavior issues. To minimize movement, sedation (Englander et al. 2015) and immobilization using restraining straps (Bandholm et al. 2009), bracing (Elder et al. 2003), or vacuum pressure (Johnson et al. 2009; Modlesky et al. 2008, 2009, 2015; Whitney et al. 2017) have been used. Dual-energy X-ray absorptiometry has also been used to assess bone architecture (Beck 2007). Currently, cortical bone architecture can be estimated in the proximal femur (Petit et al. 2002) and trabecular bone microarchitecture can be estimated in the lumbar spine (Bousson et al. 2012). However, because DXA is a two-dimensional technology, the accuracy of its bone architecture estimates is limited. Moreover, studies evaluating the distal femur, the most common fracture site in children with CP (McIvor and Samilson 1966; Presedo et al. 2007), are lacking.
Factors Contributing to Atypical Bone Growth and Development in Children with CP
Gross Motor Function and Physical Activity
The most obvious and dominant factors that contribute to the bone mass and bone architectural deficits and high fracture risk in children with the most severe forms of CP are limited weight bearing and poor motor function (Henderson et al. 2002a). While, on average, aBMD is 1.8 SD below the norm in children with CP who can ambulate with assistance (GMFCS III), it is ≥ 3.8 SD below the norm in children with CP who are nonambulatory (GMFCS IV-V) (Henderson et al. 2002a). Progressively, lower aBMD in the distal femur of children with CP is associated with lower levels of mobility (Henderson et al. 2002a).
Children with CP usually exhibit functional impairments such as spasticity, lack of dexterity, and a restricted range of motion, which may all contribute to their limited participation in daily physical activity (Bjornson et al. 2007). Compared to their typically developing peers, children with CP not only have a significantly lower physical activity but also a lower percentage of medium and high levels of activity (Bjornson et al. 2007; Johnson et al. 2009; Modlesky et al. 2009). Moreover, the physical activity performance decreases as motor function decreases. Children with CP and a GMFCS III-V have physical activity that is 70–80% lower than typically developing children (Johnson et al. 2009, Modlesky et al. 2009).
Muscle plays a very important role in bone growth. It has been proposed that forces generated during muscle contraction have a greater impact on bone than the loading provided by the gravitational force associated with weight bearing (Burr 1997; Frost 1997; Lu et al. 1997). Several studies have established a positive relationship between muscle and bone in adults and children (Bajaj et al. 2015; Lebrasseur et al. 2012; Schoenau et al. 2000; Snow-Harter et al. 1990). A significant positive relationship between thigh muscle volume and femur bone strength, but not between leg muscle volume and tibia bone strength, has been observed in 10–23 year-old individuals with bilateral spastic CP (Noble et al. 2014). In addition to the obvious biomechanical coupling of muscle and bone, muscle also acts as an endocrine organ that helps regulate bone metabolism. Muscle can produce several cytokines, including insulin-like growth factor 1 and osteonectin (Cianferotti and Brandi 2014), which have essential anabolic function in bone growth and repair. It has been demonstrated that children with CP have smaller muscles (Elder et al. 2003; Johnson et al. 2009; Modlesky et al. 2010; Shortland et al. 2002; Whitney et al. 2017), greater fat infiltration within their muscles (Johnson et al. 2009; Whitney et al. 2017), as well as a lower capacity to generate muscular force (Elder et al. 2003). Such compromise in muscle quality, coupled with the observed lower physical activity level in children with CP, suggest not only a lack of mechanical loading but also a potentially hormonal dysregulation on bone growth in this population.
In addition to the direct and indirect influence that muscles have on bone, stronger muscles can also provide mechanical advantages in fall and fracture prevention. In older adults, lower extremity muscle weakness is associated with a higher fall risk (Moreland et al. 2004). There is evidence that individuals with CP have a higher risk of falling than the general population (Ferdjallah et al. 2002; Hsue et al. 2009), which is consistent with their underdeveloped musculature and poor neuromuscular control. Increased fall risk combined with the lower aBMD, underdeveloped bone architecture and low bone strength could at least partly explain the high fracture rate in children with CP at atypical fracture sites.
Better nutritional status, as reflected by markers of growth such as height, weight, and skinfold thickness, is viewed as one of the strongest predictors of aBMD in children with CP (Henderson et al. 1995). Better nutritional status is associated with higher aBMD (Henderson et al. 2002a) and greater increases in aBMD in individuals with CP during childhood and adolescence (Henderson et al. 2005). In addition, calcium is the primary mineral stored in bone and contributes to the overall strength of bone. Vitamin D is important in facilitating the absorption of dietary calcium via the intestines (Bronner 2009). Due to oral-motor dysfunction, medication use and absorption issues associated with medication use in some children with CP, inadequate intake of calcium, vitamin D, and other nutrients important for bone growth and mineralization may occur (Henderson et al. 2002a). Some studies suggest that calcium and vitamin D supplementation can increase lumbar spine aBMD in children with severe CP (Jekovec-Vrhovsek et al. 2000). Whereas, other studies suggest that calcium intake and serum 25-hydroxyvitamin D are not correlated with aBMD (Finbraten et al. 2015; Henderson et al. 2002a) or fracture (Henderson et al. 2010) in this population. Furthermore, there is no difference in aBMD in children with CP who consume low vs. high amounts of calcium or who have low vs. normal levels of serum 25-hydroxyvitamin D (Henderson et al. 2002a). If the effect of calcium and vitamin D on bone is limited, it may be that there is insufficient mechanical stimulation to effectively use these nutrients (Duncan and Turner 1995), especially in nonambulatory children with CP. Without mechanical coupling as the first step, further steps in bone mechanotransduction may not occur. More studies are needed to determine the effect of nutrition on bone health in children with CP.
Medication use may also contribute to low bone mass and poor bone development in children with CP. In particular, there is evidence that antiepileptic drugs have a negative effect on bone in children with CP (Henderson et al. 2002a). Because epilepsy is more common in children with CP than in the general population of children (Wallace 2001), the use of antiepileptic drugs, such as phenobarbitol and diphenylhydantoin, may be a significant problem. The proposed negative effect of some antiepileptic drugs on bone is attributed to increased induction of expression of cyp24, a member of the cytochrome p450 superfamily, which may lead to greater inactivation of vitamin D (Pack 2011). Research examining the effect of antiepileptic drugs on bone in children are conflicting. In a review of the literature that included cross-sectional, cohort, case-control, and randomized controlled trials, it was concluded that carbamazepine and valproate were associated with a limited decrease in aBMD (Vestergaard 2015). The effect of antiepileptic drugs may be related to the type of medication use. Polytherapy with antiepileptic drugs has been associated with a larger decrease in aBMD than monotherapy (Vestergaard 2015) Furthermore, some studies suggest that newer antiepileptic drugs may not have a detectable effect on bone (Vohora and Anwar 2013).
It is possible that the effects of antiepileptic drugs on bone health in children with CP depends on the level of involvement of the disorder. For example, an epidemiological retrospective study that looked at the risk factors associated with fractures in children with CP suggested that there was a two-fold fracture risk increase associated with antiepileptic drug therapy for those with a GMFCS IV-V, whereas a similar relationship was not detected for those with a milder form with CP (Wort et al. 2013). However, caution is needed when interpreting observation studies that involve children with severe forms of CP, as having limited mobility itself can be associated with other issues, such as increased seizure and gastrostomy, which may have an intrinsic link to low bone mass and increased fracture risks (Henderson 2013). Therefore, the effect of antiepileptic medications on bone in children with CP remains uncertain.
Bone Health in Adults with CP
As the medical care system has improved, the life expectancy of individuals with CP has increased (Brooks et al. 2014). However, the age-related problems associated with a longer life span may include greater fracture risk for adults with CP. It is also possible that the problem in adults with CP may exceed the problem in children with CP (Sheridan 2009). Unfortunately, research studies involving bone rarely focus solely on adults with CP. One longitudinal study that included young adults with CP found a wide range of annual aBMD change with both increases and decreases observed over time (Grossberg et al. 2015). A cross-sectional study found that unlike in children, Z-scores were not related to age in adults with CP 18 to 50 years of age (Fowler et al. 2015). The findings suggest that the discrepancy in aBMD between adults with and without CP does not widen with age. On the other hand, there is evidence that the prevalence of osteoporosis in adults with CP. In addition to the limited number of studies that have directly compared the aBMD between adults with and without CP, studies examining bone architecture, bone strength and fracture patterns in adults with CP are lacking. One investigation reported that adults with spastic CP have lower femur trochanteric aBMD than adults with dyskinetic CP (Kim et al. 2015) suggesting that the type of CP should also be considered when evaluating bone health. Overall, the sparse literature suggests that more attention needs to be given to adults with CP to help us better understand their bone health and fracture risk.
A number of intervention studies suggest that regular exposure to mechanical loading can enhance the accretion of bone mineral during growth (Fuchs et al. 2001; Laing et al. 2005; Petit et al. 2002). In one of the few intervention studies focused on physical activity and bone in children with CP (Chad et al. 1999), a small group of children with spastic CP were randomly assigned to participate in a weight-bearing physical activity program or no intervention for 8 months (n = 9/group). Children in the physical activity group compared to controls demonstrated a notable increase in femoral neck bone mineral content (9.6% vs. -5.8%, p < 0.05) and volumetric BMD (5.6% vs. -6.3%, p <0.05). For nonambulatory children with CP, even standing for a longer period of time may be beneficial. A randomized controlled trial reported that nonambulatory prepubertal children with CP who stood 50% longer than their normal standing time for 9 months increased their trabecular volumetric BMD in the spine by 6% compared to controls. However, a significant change was not found in their proximal tibia (Caulton et al. 2004).
High-frequency, low-magnitude vibration also shows promise as an intervention strategy for children with CP. Some studies suggest it improves muscle mass (Gilsanz et al. 2006), strength (Leung et al. 2014; Reyes et al. 2011), and coordination (Leung et al. 2014). While low-magnitude mechanical loading is osteogenic in theory (Gilsanz et al. 2006; Leung et al. 2009; Rubin et al. 2001, 2004, Xie et al. 2006), few studies have specifically looked at its effects on bone in children with CP. Among the available studies, Wren et al. (2010) randomized children with CP (GMFCS I-IV) to either stand on a plate that emits a high-frequency, low-magnitude vibration (30 Hz, 0.3 g), or on the floor at home 10 min/d for 6 months. Children then swapped their condition for another 6 months (i.e., children who stood on the vibration plate now stood on the floor and vice versa). The vibration condition led to increases in cortical bone properties (i.e., cortical bone area and estimates of bone strength) in the tibial midshaft that were approximately double the increases observed during the floor-standing period (average increase of 16% vs. 8%). No significant changes were detected in the metaphysis of the proximal tibia or in the lumbar spine, which are highly concentrated in trabecular bone. Although studies examining the potential effect of standing, weight-bearing physical activity and mild vibration are encouraging, recent reviews of the literature (Fehlings et al. 2012; Ozel et al. 2016) concluded that there is insufficient evidence to support the recommendation of weight-bearing activities as effective interventions to improve low aBMD or to decrease fragility fractures. Intervention studies that assess larger samples sizes for a longer period of time are needed to make more definitive recommendations regarding weight bearing and bone health in children with CP.
Pharmacologically, bisphosphonates have shown excellent promise as a treatment for low aBMD in children with CP. In a randomized controlled trial, nonambulatory children with CP who received pamidronate injection showed a two- to ten-fold increase in aBMD at the distal femur and a two-fold increase in aBMD at the lumbar spine relative to placebo controls 6 months after treatment (Henderson et al. 2002b). In another study that involved 23 nonambulatory children with spastic CP, low-dose pamidronate was given intravenously. The 12-month intervention yielded a significant increase in aBMD in both the lumbar spine and femoral neck (Plotkin 2006). A recent investigation also found that orally administering alendronate (1 mg/kg/week) increased lumbar spine aBMD in children with quadriplegic CP without inducing side effects (Paksu et al. 2012). Based on the available published studies, a recent review concluded that there was “probable” evidence that bisphosphonates are effective at improving low aBMD in children with CP (Ozel et al. 2016).
Bisphosphonates also show promise in the reduction of fragility fractures in children with CP. For example, one study reported that the percentage of fractures in children with CP dropped from 31% to 13% per year after 13.6 months of pamidrotate treatment (Bachrach et al. 2010). However, due to the limited number of studies available, a recent review concluded that there was only “possible” evidence that bisphosphonates are effective at reducing fragility fractures in children with CP (Ozel et al. 2016). To date, more definitive studies examining the effect of bisphosphonates on fragility fractures are needed. Studies are also needed to determine if the potential effect of bisphosphonates on bone fragility fractures in children with CP is mediated, at least in part, by changes in bone architecture and increases in bone strength.
Poor feeding and undernutrition is common in children with CP (Fung et al. 2002; Kim et al. 2018). Recent reviews of the literature concluded that there is “possible” evidence for calcium and vitamin D to improve low aBMD in children with CP (Fehlings et al. 2012, Ozel et al. 2016). However, there is inadequate evidence to support the use of calcium and vitamin D supplementation to decrease fragility fractures in children with CP. Due to the possible effectiveness of calcium and vitamin D, their good safety record, and their existing recommendation for children, an approach that includes vitamin D supplementation and adequate calcium intake through the diet is viewed as good clinical practice (Fehlings et al. 2012). However, further research is necessary (Ozel et al. 2016).
Children with CP have significant deficits in bone health, as demonstrated by low aBMD, low bone mineral content, less developed bone architecture, and very low bone strength. Although the degree of the deficits are greatest in nonambulatory children with the most severe forms of CP, significant deficits are present even in ambulatory children with milder forms of CP. The limited data available suggest that bone health is also compromised in adults. However, the extent of the deficit and whether it deteriorates at a greater rate than observed with typical aging is unknown. Treatment strategies that include standing, weight-bearing, physical activity, high-frequency, low-magnitude vibration, bisphosphonates, and calcium/vitamin D may improve the bone deficits in children with CP. However, more studies are needed to determine the most effective treatments, the timing of intervention, and their short- and long-term benefits.
- Bousson V, Bras A, Roqueplan F, Kang Y, Mitton D, Kolta S, Bergot C, Skalli W, Vicaut E, Kalender W, Engelke K, Laredo JD (2006) Volumetric quantitative computed tomography of the proximal femur: relationships linking geometric and densitometric variables to bone strength. Role for compact bone. Osteoporosis Int 17(6):855–864CrossRefGoogle Scholar
- Fehlings D, Switzer L, Agarwal P, Wong C, Sochett E, Stevenson R, Sonnenberg L, Smile S, Young E, Huber J, Milo-Manson G, Abu Kuwaik G, Gaebler D (2012) Informing evidence-based clinical practice guidelines for children with cerebral palsy at risk of osteoporosis: a systematic review. Dev Med Child Neurol 54(2):106–116PubMedCrossRefGoogle Scholar
- Fung EB, Samson-Fang L, Stallings VA, Conaway M, Liptak G, Henderson RC, Worley G, O'donnell M, Calvert R, Rosenbaum P, Chumlea W, Stevenson RD (2002) Feeding dysfunction is associated with poor growth and health status in children with cerebral palsy. J Am Diet Assoc 102(3):361–373PubMedCrossRefGoogle Scholar
- Henderson RC, Berglund LM, May R, Zemel BS, Grossberg RI, Johnson J, Plotkin H, Stevenson RD, Szalay E, Wong B, Kecskemethy HH, Harcke HT (2010) The relationship between fractures and DXA measures of BMD in the distal femur of children and adolescents with cerebral palsy or muscular dystrophy. J Bone Miner Res 25(3):520–526PubMedCrossRefGoogle Scholar
- Lebrasseur NK, Achenbach SJ, Melton LJ 3rd, Amin S, Khosla S (2012) Skeletal muscle mass is associated with bone geometry and microstructure and serum insulin-like growth factor binding protein-2 levels in adult women and men. J Bone Miner Res 27(10):2159–2169PubMedPubMedCentralCrossRefGoogle Scholar
- Leung KS, Li CY, Tse YK, Choy TK, Leung PC, Hung VW, Chan SY, Leung AH, Cheung WH (2014) Effects of 18-month low-magnitude high-frequency vibration on fall rate and fracture risks in 710 community elderly--a cluster-randomized controlled trial. Osteoporos Int 25(6):1785–1795PubMedCrossRefGoogle Scholar
- Marks SC, Odgren PR (2002) Structure and development of the skeleton. In: Principles of bone biology, 2nd edn. Academic Press, San DiegoGoogle Scholar
- Martin RB, Burr DB, Sharkey NA, Fyhrie DP (2015) Skeletal tissue mechanics. Springer, New YorkGoogle Scholar
- Modlesky CM, Cavaiola ML, Smith JJ, Rowe DA, Johnson DL, Miller F (2010) A DXA-based mathematical model predicts midthigh muscle mass from magnetic resonance imaging in typically developing children but not in those with quadriplegic cerebral palsy. J Nutr 140(12):2260–2265PubMedPubMedCentralCrossRefGoogle Scholar
- Modlesky CM, Whitney DG, Singh H, Barbe MF, Kirby JT, Miller F (2015) Underdevelopment of trabecular bone microarchitecture in the distal femur of nonambulatory children with cerebral palsy becomes more pronounced with distance from the growth plate. Osteoporosis Int 26(2):505–512CrossRefGoogle Scholar
- Vohora D, Anwar MJ (2013) Phenytoin and sodium valproate (but not levetiracetam) induces bone loss in swiss albino female mice, prevention by raloxifene, role of estradiol/TGF beta3 pathway. Osteoporosis Int 24:S611–S611Google Scholar
- Woodhead HJ, Kemp AF, Blimkie CJR, Briody JN, Duncan CS, Thompson M, Lam A, Howman-Giles R, Cowell CT (2001) Measurement of midfemoral shaft geometry: repeatability and accuracy using magnetic resonance imaging and dual-energy X-ray absorptiometry. J Bone Miner Res 16(12):2251–2259PubMedCrossRefGoogle Scholar
- Wren TAL, Kalkwarf HJ, Zemel BS, Lappe JM, Oberfield S, Shepherd JA, Winer KK, Gilsanz V (2014) Longitudinal tracking of dual-energy X-ray absorptiometry bone measures over 6 years in children and adolescents: persistence of low bone mass to maturity. J Pediatr 164(6):1280–1285PubMedPubMedCentralCrossRefGoogle Scholar