Complicated muscle-bone interactions in children with cerebral palsy

Abstract

Purpose of review:

The goal of this review is to highlight the deficits in muscle and bone in children with cerebral palsy (CP), discuss the muscle-bone relationship in the CP population and identify muscle-based intervention strategies that may stimulate an improvement in their bone development.

Recent findings:

The latest research suggests that muscle and bone are both severely underdeveloped and weak in children with CP, even in ambulatory children with mild forms of the disorder. The small and low-performing muscles and limited participation in physical activity are likely the major contributors to the poor bone development in children with CP. However, the muscle-bone relationship may be complicated by other factors, such as a high degree of fat and collagen infiltration of muscle, atypical muscle activation, and muscle spasticity. Muscle-based interventions, such as resistance training, vibration and nutritional supplementation, have the potential to improve bone development in children with CP, especially if they are initiated before puberty.

Summary:

Studies are needed to identify the muscle-related factors with the greatest influence on bone development in children with CP. Identifying treatment strategies that capitalize on the relationship between muscle and bone, while also improving balance, coordination and physical activity participation, is an important step toward increasing bone strength and minimizing fractures in children with CP.

Introduction

Skeletal muscle and bone are key tissues in the body that possess independent and complementary roles. Skeletal muscle serves as a reservoir for amino acids (1) and carbohydrate (2); whereas, bone serves as a reservoir for calcium (3) and other micronutrients. Together, muscle and bone are needed for physical movement, the protection of vital organs and the regulation of metabolic activity.

Cerebral palsy (CP) is the most common cause of childhood-onset physical disability in most countries (4), affecting approximately 17 million people worldwide (5). It is a heterogeneous disorder of movement and posture resulting from an injury to or a malformation of the motor areas of the developing brain (5, 6). The associated causes of the brain injuries or malformations acquired during the prenatal, perinatal, or postnatal periods, include trauma, hemorrhage, stroke, hypoxia-ischemia and infection. Moreover, emerging evidence suggests that genetic variants may play an important role in the pathogenesis of CP (7–9).

In terms of clinical manifestations, individuals with the mildest form of CP are able to walk, run and jump, although to a lesser degree than individuals with typical development. Individuals with the most severe form of CP are nonambulatory, even with assistance (10). The disorder is also categorized into clinical subtypes based on the limbs and sides affected. For example, one side of the body is affected in individuals with unilateral or hemiplegic CP, the lower extremities are primarily affected in individuals with bilateral or diplegic CP, and the upper and lower extremities are primarily affected in individuals with tetraplegic or quadriplegic CP.

The goal of this review is to highlight the marked deficits in muscle and bone in children with CP, discuss the muscle-bone relationship in the CP population and identify muscle-based intervention strategies that may stimulate an improvement in their bone development.

Muscle deficits in children with CP

Muscle strength and other aspects of muscle performance:

The low level of motor function in individuals with CP is related to their poor muscle performance. One aspect of muscle performance is muscle strength. Muscle weakness is a hallmark of CP (11–13), with the most profound level of weakness at high concentric velocities (14) and in nonambulatory children with the most severe forms of the disorder. However, muscle weakness is present even in ambulatory children with milder CP. For example, maximum voluntary isometric contractions 56 % lower for the knee extensor forces and 73 % lower for the plantarflexors have been reported in children with spastic diplegic CP compared to children without CP (12). Moreover, the strength deficit is greater distally than proximally (13) and may become progressively more pronounced with age during childhood and adolescence (15). The deficit in muscle performance in individuals with CP is not limited to muscle strength. Muscle power (16) and rate of force development (17) are also deficient, and the level of deficiency is even greater than that of muscle strength.

A major factor that contributes to reduced muscle performance in children with CP is their restricted muscle size (11, 18–22). However, other factors may also contribute to their muscle performance issues, such as the fat (20, 22, 23) and collagen (24) infiltration of muscle, impaired muscle activation (11, 12, 25) and muscle spasticity. These other factors may also explain the reduced force production relative to muscle size in children with CP when compared to typically developing children (11).

Muscle size and architecture:

Muscles are considerably smaller in children with CP than in typically developing children (11, 18–22). Midthigh muscle mass (21) and cross-sectional area (CSA) (20) are approximately 50 % lower in children with CP who are unable to ambulate without assistance than in typically developing children. Although to a lesser extent, ambulatory children with CP also have muscle size deficits. For example, a recent study reported 39 % lower midleg muscle volume in ambulatory children with spastic CP compared to typically developing children matched to children with CP for sex, age and race and not different from the 50^th sex- and age-based percentiles for height, body mass and body mass index (22). There is also evidence indicating a more substantial muscle size deficit in the affected lower limb of children with hemiplegic CP compared to the lower limbs of children with diplegic CP (26). However, whether there is a deficit in muscle size in the unaffected limb of children with hemiplegic CP is unclear. In addition to muscle mass, CSA and volume, muscle belly length is compromised in children with CP. Medial gastrocnemius muscle belly length normalized to fibular length is 12 to 19 % lower in children with spastic diplegic CP compared to typically developing children (18, 27). The muscle size deficit in children with CP is also more profound distally than proximally. For example, although muscle volume in the thigh (i.e., between the hip and knee) and leg (i.e., between the knee and ankle) are both lower in the affected vs. the unaffected side in adolescents and young adults with hemiplegic CP, the discrepancy is greater in the leg (28, 29).

Studies examining the effect of CP on measures of muscle architecture, such as fascicle length and fascicle (i.e., pennation) angle, have yielded inconsistent results. When compared to typically developing children, no group difference (19, 26, 27, 30, 31), lower (27) and higher (26) fascicle length and no group difference (13, 20, 24, 27), lower (26) and higher (27) fascicle angle have been reported in children with CP. The conflicting studies may be related to inconsistency in measurement procedures used in the different studies, such as differences in the joint angle at which muscle fascicle length and angle were measured. In addition, a reduction in fascicle length has been reported after surgical recession of the gastrocnemius (31), a common procedure in children with spastic CP. However, it is possible that there are other unexplained contributors.

Fat and collagen infiltration of muscle:

Cross-sectional studies have reported that subcutaneous fat in children with CP is not different from that in typically developing children (20, 22). On the other hand, intermuscular fat is 2–3 fold higher in the thigh of children with quadriplegic CP than in typically developing children and the degree of intermuscular fat is inversely related to physical activity (20). Higher levels of intramuscular and intermuscular fat have also been reported in the leg muscles of ambulatory children with spastic CP compared to typically developing children matched for sex, age and race (22). Furthermore, higher intramuscular fat in the leg muscles has been observed in young adults with spastic CP. The most pronounced infiltration of fat was in the soleus. Moreover, there was a higher concentration of intramuscular fat in adults who needed assistance to ambulate as compared to adults who could ambulate without assistance (23). A high degree of fatty infiltration in skeletal muscle (i.e., myosteatosis) is associated with aging, glucocorticoid treatment, unloading or disuse, sex steroid deficiency and leptin deficiency (32). It is also associated with chronic diseases, such as type 2 diabetes mellitus (33), cardiovascular disease (34) and osteoporosis (35). In addition to the fatty infiltration of muscle, there is evidence of a greater collagen concentration in the muscle of children with CP with increasing levels of spasticity (24).

Muscle activation:

An inability to maximally activate muscles (11, 12) and a higher level of coactivation of antagonist muscles during voluntary muscle contraction (11, 12, 25) have been reported in children with CP relative to typically developing children. Reasons for this poor muscle activation may be related to impaired motor unit recruitment and/or firing rate modulation (36, 37). It is also plausible that dysmorphic neuromuscular junctions (38) limit the transmission of action potentials needed for muscle contraction. As with muscle size, the deficits in muscle activation of individuals with CP may not be limited to the side that is primarily affected. A lower maximum voluntary contraction and maximum voluntary contraction relative to anatomical CSA, a surrogate of specific tension, has been reported in the plantar flexors and dorsiflexors of the affected and unaffected limbs in children with hemiplegic CP (11).

Muscle spasticity:

Another factor that may limit muscle performance is muscle spasticity, which is a velocity-dependent increase in muscle tone due to hyperexcitability of the stretch reflex. It is present in ~75 to 80 % of children with CP (4, 39). Muscle spasticity limits range of motion, is mildly related to motor function (40) and may reduce participation in physical activity. Muscle spasticity is also associated with the development of muscle contractures (41). There is evidence that the muscles of children with CP with contractures have fewer satellite cells (42), a stiffer extracellular matrix and a greater in vivo sarcomere length (43) than the muscles of typically developing children.

Bone deficits in children with CP

Bone strength and fragility fractures:

In addition to having very weak and low-performing muscles, individuals with CP have very weak bones. This is demonstrated by 60 to 70 % lower estimates of bone strength in the midfemur of children with CP who are unable to ambulate independently than observed in typically developing children. Although not as extensive, ambulatory children with CP also have considerable bone strength deficits. Estimates of bone strength are ~ 30 – 40 % lower in ambulatory children with CP than in typically developing children matched for sex, age and race and who are not different from the 50^th percentile for height, body mass and BMI. The bone weakness in children with CP is associated with a high rate of fragility fractures, especially in nonambulatory children (44). There is evidence that nearly half of all fractures in children with CP occur in the femur and that the distal femur is the most common fracture site (44). On the other hand, only 2 % of fractures occur in the femur of the general population of children (45). Several factors contribute to bone weakness in individuals with CP, such as restricted bone size (22, 46, 47), underdeveloped cortical bone architecture (22, 46, 47), underdeveloped trabecular bone architecture (48, 49), low areal bone mineral density (aBMD), low bone mineral content (48, 50, 51) and elevated fat infiltration of bone marrow (22).

Bone size and architecture:

As with muscle, bone growth and development are restricted in children with CP. This is demonstrated by reports of children with CP having shorter bones, which contribute to their shorter stature (47, 52). In addition, compared to matched typically developing children, children with CP who are unable to ambulate without an assistive device have a shaft, cortex and medullary cavity that are 27–43 % thinner and have 51–55 % less volume at the midfemur (47). Similar to the deficits in bone strength, the magnitude of architectural deficits is not as extensive in ambulatory children as in nonambulatory children with CP. However, the deficits are notable. Compared to matched typically developing children, ambulatory children with CP have a 15 % thinner shaft, a 15–30 % thinner cortex and a 30 % lower cortical volume at the midtibia (22).

In addition to having deficits in bone size and cortical bone architecture, children with CP also have a considerable deficit in trabecular bone architecture. Their bones compared to the bones of typically developing children have fewer and thinner trabeculae. Specifically, 30 % lower trabecular bone volume to total volume, ~ 20 % fewer trabeculae and 12 % thinner trabeculae have been observed in the distal femur immediately above the growth plate (48, 49). Moreover, the deficit becomes increasingly more pronounced with greater distance from the growth plate (49). The deficit in trabecular bone microarchitecture in the distal femur of children with CP, which is highlighted in Figure 1, is consistent with their high rate of fragility fractures at this bone site (44).

Figure 1.

The figure summarizes the deficit in trabecular bone microarchitecture observed in children with cerebral palsy (CP). Trabecular bone microarchitecture was assessed in the lateral distal femur immediately above the growth plate (A), as indicated by the dotted rectangle, using magnetic resonance imaging. Binarized magnetic resonance images show the remarkable underdevelopment of trabecular bone microarchitecture in a nonambulatory boy with CP (B) when compared to typically developing boy of the same age and near the 50^th percentile for height and body mass (C). The images were modified from Modlesky et al. (49).

Areal bone mineral density and mineral content:

Children with CP also have low aBMD and bone mineral content measured by dual-energy X-ray absorptiometry (48, 50, 51). Although these bone indices are lower in the appendicular skeleton (i.e., distal femur) and the lumbar spine of children with CP when compared to typically developing children, the distal femur exhibits a more pronounced deficit (50). The compromise in distal femur aBMD is also greatest in the children with the most severe forms of CP (i.e., unable to ambulate even with assistance) (50). The low aBMD and low bone mineral content are primarily due to the smaller bones of children with CP compared to typically developing children (50). One study suggests that cortical volumetric bone mineral density, a measure not influenced by bone size, is not compromised and may even be higher in children with CP (46). However, more studies are needed to confirm these findings.

Fat infiltration of bone marrow:

Another factor that may contribute to the low bone strength and increased bone fragility in children with CP is the high concentration of fat in their bone marrow (22). There is evidence that bone marrow fat is negatively related to mechanical loading (54, 55) and bone architecture (56–63). Increased bone marrow fat is associated with reduced glucose uptake (64) and other markers of type 2 diabetes mellitus (64–66). Therefore, bone marrow fat may be an important modulator of skeletal homeostasis and may contribute to systemic energy metabolism. Unfortunately, studies examining the influence of bone marrow fat on skeletal and overall metabolic health during human growth and development are lacking, especially studies focused on children with CP.

Muscle and bone interactions in children with CP

Muscle and bone early growth and development:

Muscle and bone have a very close connection that begins in utero (67). The development and maintenance of skeletal muscle and bone are regulated by morphogens and growth factors, many of which overlap between the two tissues, such as Wnts, Hedgehog, Notch, fibroblast growth factors, insulin-like growth factor 1 and transforming growth factor beta (68). The relationship between muscle and bone extends postnatally. This is clearly demonstrated by cross-sectional studies exhibiting a strong and consistent relationship between muscle and bone in healthy individuals (69, 70). Because of the close tie between muscle and bone, it is logical to expect that the deficits in muscle (Fig. 2) and the deficits in bone (Fig. 1 and 2) in children with CP are related.

Figure 2.

Raw T1-weighted magnetic resonance images of the midtibia and the surrounding leg musculature and adipose tissue taken from an ambulatory boy with mild CP (A) and a typically developing boy not different in age or race (B). The images demonstrate the marked deficit in cortical bone architecture (small arrow indicating a thin tibia cortex) and muscle size (gray) and the high infiltration of fat within (intramuscular; medium arrow) and adipose tissue around (intermuscular; large arrow) the musculature. The figure was modified from Whitney et al. (22).

Mechanical influence of muscle on bone:

The most obvious reason for the concomitant deficits in muscle and bone in children with CP is linked to the limited mechanical stimulation of muscle on bone. According to Wolff’s law (71) and the “Mechanostat” theory proposed by Frost (72), bone development is heavily influenced by its mechanical loading environment. Muscles attach to the skeleton and exert force directly to the bone. Consequently, force generated by skeletal muscle contraction has been viewed as a primary stimulus for bone mechanical signaling (73). The underlying mechanism by which the signaling is believed to occur is via a response to strain magnitude (53), rate (74), and frequency (75) and/or changes in fluid flow within the bone (76). The magnitude of strain on the bones of children with CP is likely low due to their small (11, 18–22) and weak (11) muscles. In addition, the rate of strain on the bones of children with CP is probably low due to the low power generation (16) and slow rate of force development (17) by their muscles. Some of the poor muscle performance is linked to the high degree of fat (20, 22) and collagen (24) infiltration of the muscles. When considering that the deficits in muscle size, quality and performance in children with CP are coupled with significant balance and coordination issues (77–80), it is not surprising that very low participation in physical activity results (20, 22, 47, 81, 82) and leads to less strain frequency and mechanical stimulation of bone. Physical activity is 50 to 80 % lower in children with CP than in typically developing peers (20, 22, 47, 81, 82).

The notion that mechanical stimulation closely connects muscle and bone in individuals with CP is supported by a study of adolescents and young adults with and without CP (83). The lower femur cortical CSA, cortical thickness and bone-bending strength in the group with CP were significantly related to the lower volume of the surrounding thigh musculature. However, another interesting observation was that tibia cortical CSA was not significantly related to the volume of the surrounding leg musculature in the combined sample of children with CP and typically developing children (p = 0.166). The conflicting findings suggest that the relationship between muscle and bone is complex. The degree of the complexity may be even greater in individuals with CP. Their atypical muscle activation and movement patterns, as well as muscle spasticity, may lead to a bone or bone site being influenced by a muscle, or a combination of muscles, that is different from typically developing children. This idea is supported by the bone deformities that emerge during the growth of children with CP (84). It is also supported by a study by Ziv et al. (85), in which the muscle-bone relationship became disrupted in a group of young mice who developed spasticity. While the naturally occurring increases in muscle (gastrocnemius) and bone (tibia) length were both restricted in mice who developed spasticity, the restriction was much greater in muscle length. Moreover, although low aBMD has been observed in adults with spastic and dyskinetic CP, the deficit is greater in adults with the spastic subtype (86).

Nonmechanical influence of muscle on bone:

There are several lines of evidence supporting a nonmechanical role of muscle in bone growth and development, which may be applicable to children with CP. Using myostatin-null mice model, Hamrick et al. (87) showed that increased muscle volume is associated with favorable bone growth, despite the lack of a proportionate increase in muscle strength. In addition, bone fractures due to orthopedic injury exhibit improved healing when they are covered with muscle flaps (88, 89). Moreover, muscle implanted beside the periosteum can stimulate bone formation (90). Conversely, the healing of bone with a defect can be impaired if the surrounding muscle is damaged or experiences trauma (91, 92). The studies mentioned above (87–92) suggest that muscle contributes to bone development in ways other than just mechanical stimulation. Endocrine factors are believed to play an important role in this alternate process. Muscle, as an endocrine organ, can secrete insulin-like growth factor 1 (93, 94) and fibroblast growth factor 2 (95, 96), both of which exhibit osteogenic effects. Other myokines may provide an osteogenic effect as well (97–99), and some may have anabolic effects on ligaments and cartilage, which can help enhance the quality of bone (97). It is possible that the high degree of fat infiltration in the muscles of individuals with CP (20, 22, 23) adversely influence their interaction with bone. In addition to causing a reduced force generation capacity (100), fat infiltration of muscle decreases insulin sensitivity, which impairs the capacity for protein synthesis in skeletal muscle (101).

Potential effect of common treatments:

Treatments commonly used to manage the secondary complications associated with CP may alter the muscle-bone relationship. For example, botulinum toxin A is often used to treat muscle spasticity in children with CP. A number of studies have shown that botulinum toxin A injected into the muscles of healthy animals leads to a marked deterioration of the treated muscle and the bone to which it attaches (102–104). Therefore, it is plausible that botulinum toxin A would cause a similar musculoskeletal decline in children with CP. Nevertheless, there is evidence that a spastic gastrocnemius muscle treated with botulinum toxin declines only modestly in volume, and the drop is offset by an increase in the volume of the soleus and is not accompanied by change in physical function (105). Other procedures may also interfere with the muscle-bone relationship in individuals with CP, such as surgical recession of the gastrocnemius, which has been shown to suppress muscle belly length (18). However, studies that examine the effect of surgical recession on bone and the muscle-bone relationship in individuals with CP are needed.

Potential effect of muscle-based interventions on bone

Resistance training and physical activity:

A number of treatments could potentially aid in the growth and development of muscle and bone in children with CP. Resistance training and intense physical activity have been encouraged as a part of the treatment plan for children with CP (106). This recommendation has been supported by a growing number of studies, with resistance training that emphasizes high-velocity movement exhibiting the greatest promise. For example, a randomized controlled trial found that 6 weeks of functional resistance training led to increases in quadriceps femoris muscle thickness, rectus femoris muscle CSA, and medial gastrocnemius pennation angle in children with spastic CP who were able to ambulate with or without assistance (107). Moreover, a small randomized controlled trial by Moreau et al. (16) found that velocity-based resistance training vs. traditional resistance training led to greater increases in velocity of movement, muscle power and walking performance (p < 0.10). Although the currently available literature is promising, more high-quality studies with larger sample sizes are needed to confirm the findings (108), to aid in developing specific training guidelines and to determine whether a concomitant effect on bone would result.

Vibration:

High-frequency, low-magnitude vibration is also a promising treatment for muscle and bone deficits in children with CP. It has been shown to increase the size of some muscles, such as the tibialis anterior and soleus (109), but not others, such as the gastrocnemius (109). There is also evidence that it can increase cortical bone properties in the midtibia (110). Moreover, there is evidence that vibration can be effective whether it is delivered to the whole body (110) or to a specific region of the body (111). Nonetheless, more studies are needed to confirm the findings and to determine the frequency, intensity (i.e., signal frequency and displacement), and duration of the most effective treatment. More studies are also needed to determine the site-specific effects of vibration. There is evidence that the transmission of a floor-based vibration is amplified at the distal tibia, but dampened at the distal femur (112), which is the primary fracture site in children with CP (44). In addition, there is evidence that vibration can improve balance (113), which may lead to increases in physical activity and a subsequent positive effect on muscle and bone. However, again, studies are needed to confirm this idea.

Nutrition:

Poor nutritional intake is associated with stunted growth in children with CP (114–116). Consequently, nutritional interventions may improve the muscle and bone deficits in children with CP (117). Protein, essential amino acids and vitamin D show the most promise, especially if they are coupled with resistance training or physical activity (117), but studies are needed to assess their effectiveness in children with CP.

Timing of interventions:

When compared to other conditions that have a profound effect on motor function, such as spinal cord injury and stroke, the bone strength deficit in children with CP is remarkable. The bone strength deficit in children with CP who are unable to ambulate independently is 3 to 5 times greater than the bone strength deficit observed in adults with spinal cord injury (118) and approximately 15 times greater than the strength deficit in adults with a stroke (119). The reason for the more dramatic strength deficit in children with CP than observed in spinal cord injury and stroke is likely the timing of the injury. Most spinal cord injuries occur during late adolescence and young adulthood (120) and most strokes occur during mid to late adulthood (121) when the prime period for bone growth and development has already passed. On the other hand, the injury or malformation associated with CP occurs sometime between in utero and the first two years after birth. Therefore, bone growth (i.e., length, endosteal surface and periosteal surface) is restricted and the bone does not reach its full potential of size and strength.

Considering the timing of the brain injury or malformation associated with CP and the dramatic effect on the musculoskeletal system of individuals with CP, early intervention that facilitates improvements in muscle size, quality and performance and the action of muscle on bone via increased participation in physical activity is critical. Specifically, muscle-based interventions, especially those that include mechanical loading and are initiated before puberty, are more likely to create positive changes in bone (122–125) that are sustained throughout adulthood (125), as suggested in Figure 3. The idea is supported by cross-sectional studies demonstrating marked side-to-side differences in bone properties of racquet sport players (122). It is also supported by intervention studies demonstrating much more substantial increases in bone mass in children (123, 124) that are partially sustained in adolescence (126) compared to modest changes in adults (127).

Figure 3.

A raw coronal magnetic resonance image of thigh muscles and femurs (A) and segmented images showing the smaller and thinner cortical bone of a nonambulatory boy with CP (black ring; B) compared to the cortical bone of a typically developing boy not different in age or race (C). The images were generated from participants in Modlesky et al. (47). Also shown (D) are the theoretical changes in bone in: 1) typically developing children, 2) children with CP involved in a muscle-based intervention (e.g., resistance training) and not receiving a common treatment that adversely affects muscles, and 3) children with CP not involved in a muscle-based intervention and receiving a common treatment that adversely affects muscle. The figure was modified from Modlesky and Lewis (125).

Conclusion

There is a marked underdevelopment of muscle and bone in children with CP. The small and low-performing muscles of children with CP coupled with their limited physical activity participation are obvious contributors to their bone weakness and high incidence of fragility fractures. However, there may be additional contributing factors that complicate the muscle-bone relationship in children with CP, such as a high degree of fat and collagen infiltration of muscle, atypical muscle activation, and muscle spasticity. Studies should be designed to test promising interventions that can promote improvements in bone through muscle-based mechanical and/or nonmechanical stimulation and capitalize on the interrelationship between muscle and bone, while also improving the coordination of movement and physical activity participation of children with CP. Although the earliest possible intervention is encouraged, the effect of intervention at all stages of development should be evaluated.