2.1.2.1 Skeletal muscle: excitability and contraction

The main function of skeletal muscle is to provide movement, which is facilitated through its properties of excitability, contractility, elasticity and extensibility (see Box 2.2). Excitability and contractility (capacity for contraction) are described in this section. Skeletal muscles and connective tissues share the properties of elasticity and extensibility which are addressed in Section 2.1.2.3.

Skeletal muscles facilitate limb movement by adjusting their length and tension through the contraction of muscle fibres which are organised into motor units. Each motor unit consists of a single alpha motor neuron (with its cell body located in either the brainstem or ventral horn of the spinal cord), its axon (the long fibre of the neuron that conducts nerve impulses and synapses with the muscle fibre at the neuromuscular junction), and the muscle fibres that it innervates. The muscle fibres within each motor unit are of the same metabolic type, and the number of fibres associated with each motor unit depends on where the muscle is located and the nature of its action. For example, in the hand where finer movements are required, a motor unit will be associated with fewer fibres (it is a small motor unit) and will generate less force, while in a large, more powerful muscle it may be associated with thousands of fibres (a large motor unit) and will generate greater force on contraction [8].

There are different ways of classifying motor units. A common approach is to classify them according to the speed of contraction (or twitch) and the fatigability (or endurance) of their muscle fibres (Table 2.2). Classified in this way, there are three motor unit types: type 1 or slow-contracting, slow-fatigable (S), type 2A or fast-contracting, fatigue-resistant (FR), and type 2B/2X or fast-contracting, fast-fatigable (FF) [14–17].

The production of muscle force depends on a variety of neuromuscular factors (see Table 2.3), including the number and type of motor units recruited in a contraction and their firing rates. Muscle strength or force increases when the number and/or the firing rates of already-activated motor units increases. Changes in firing rates allow adjustment of force production, with different motor units having an optimal range of firing over which tension increases. An orderly recruitment pattern also occurs according to the amount of force required for task completion. Low force-producing motor units are recruited first (type 1), with higher force-producing motor units recruited as necessary (type 2A then type 2B/2X). Thus, the succession of recruitment of motor units provides a smooth increase in tension development [18, 19]. In general, the muscles of the upper limbs contain more type 2B than other types of fibres, at least in their superficial areas, with an increase in type 1 fibres in their deeper areas. This means that the muscles are able to respond with fast contraction, which is then sustained by recruitment of less fatigable fibres [20].

^a See Figure 2.1.

^b Fascicles are groups of muscle fibres.

^c Muscle belly is the length of the whole muscle (i.e. many fascicles).

Sarcomeres are the contractile part of muscle, found within the muscle fibres and grouped together into fascicles. Each sarcomere is made up of two types of overlapping muscle proteins or filaments: myosin (thick filaments) and actin (thin filaments). During contraction, the actin filaments slide toward one another past the myosin filaments, breaking and reforming cross-bridges in a process called cross-bridge attachment (see Figure 2.1). The amount of tension that can be generated by a muscle during contraction depends on the number of cross-bridges that form, or alternately, the amount of overlap between the actin and myosin filaments. The best position for the muscle to develop maximal contraction and force generation (that is, the optimal range of overlap for the filaments), is when the muscle is at its resting length. If the muscle is held in a shortened position, then the thin actin filaments are already close together, and are unable to overlap the thick myosin filaments any further because the overlapping zone is already as large as possible. If the muscle is held in a lengthened position the thick and thin filaments may be too far apart to and the zone of overlap will be small. In both these situations the potential of the muscle to contract and generate force is reduced. Skeletal muscle cells can contract until they shorten by about 30% [21].

Titin (or connectin) is another protein molecule found in the sarcomere. It is a long, coiled molecule that is wrapped around the myosin filaments and attached to the Z-line that forms the borders of each sarcomere. Titin behaves like a spring, returning the sarcomere to its original length following stretch [22, 23]. It has been suggested that different muscle groups may contain different forms of titin which influence the tension and elastic limits within sarcomeres, meaning that some muscles may have more capacity for stretch or extensibility than others [24].

2.1.2.2 Connective tissue: strength and flexibility

Ligaments, tendons, joint capsules, tendon sheaths, cartilage and other related structures are collectively described as connective or fascial tissues. The structural fibres of connective tissues (collagen, elastin and fibrin) are held together by chemical bonds and are surrounded by a filler gel that lubricates the fibres. Collagen is a fibrous protein that gives connective tissues their strength and flexibility. Connective tissues differ in terms of the type (more than 20 types of collagen have been identified), amount, density and alignment of their collagen fibres [29]. For example, the collagen fibres of dense connective tissues (bone, tendons and ligaments) are tightly packed and mostly aligned in one direction. In contrast, loose connective tissues (in muscle, joint capsules and fascia) have irregular or crisscrossed collagen fibre alignment and are more flexible [30, 31].

Tendons attach muscle to bone and transmit mechanical forces through the muscle–tendon unit, allowing for joint movement [32]. They are integrated into the neuromuscular system through sensory receptors (Golgi tendon organs) that are embedded in the muscle–tendon junction, sending information about changes in muscle tension to the CNS [8, 33]. Ligaments connect bones or cartilage to one another and, together with joint capsules, provide stability to joints [30, 34].

Articular or hyaline cartilage is another form of connective tissue. It is flexible and functions to distribute loads within the joint and to minimise friction between articulating joint surfaces. Cartilage differs from other connective tissues in that it does not contain blood vessels. This means that it grows more slowly and is also slower to repair itself if damaged [35].

Bone is a specialised connective tissue that contains tightly packed, parallel collagen fibres and inorganic minerals (mainly calcium phosphate) which provide rigidity. Bone has many blood vessels and is therefore able to repair itself relatively quickly. It is also highly adaptable to the mechanical demands that are placed on it. For example, changes in density are common in response to either disuse or increased use [34, 36].

2.1.2.3 Muscle and connective tissue: elastic, plastic, viscous and viscoelastic properties

Soft tissues show different responses to various deforming forces, whether compressive, tensile (stretch) or shear (sliding). The different responses to such forces include elasticity, plasticity, viscosity and viscoelasticity (see Box 2.3) [12, 30].

Muscles and tendons are both characterised by elasticity, although this is due to different mechanisms. The elastic property of tendons is mainly due to the ‘crimped’ (zigzag, or wavy) structure of their collagen fibrils, which straighten out as tension is applied [22, 32]. The elasticity of muscles is thought to be due to several possible mechanisms. These include (i) the small amount of the protein elastin in the intramuscular connective tissue that surrounds muscle fibres, (ii) weakly formed actin and myosin cross-bridges in the contractile sarcomere and (iii) the coiled protein, titin, also found in the sarcomere (see Section 2.1.2.1). Titin is described as the structure most likely to be responsible for muscle’s elastic response [22, 23].

In addition to their elastic properties, connective tissues display viscoelasticity when their load (or stretch) is maintained. This is related to the combination of their elastic properties and their fluid or gel components (these components are viscous, having a glutinous consistency). When connective tissues are subjected to a slow stretch and are then held at their new length with the joint at a stable angle, resistance in the tissues gradually reduces to a steady level (this reduction is called stress relaxation), and less force is needed to hold the joint angle stable. If the tissues are subjected to a slow but constant force (torque) during stretch, the tissues will continue to elongate (this elongation is called creep) and the joint angle increases [12, 22]. When resting muscle is subjected to small, low-load passive stretch it shows a very high resistance to that stretch, called short range stiffness. The bonds between myosin and actin filaments are proposed to be responsible for this stiffness [37]. However, if the muscle is held in a constant, stable position, the resistance reduces and stress relaxation will occur as described above for connective tissues. If, after stretch and viscous deformation (creep), the muscle is shortened and returned to its original position at which the stretch was begun, resistance will be found to have returned as well. This return of resistance is called stress recovery [12, 22].

2.1.2.4 Muscle tone

Muscle tone is described as the constant state of mild tension observed in muscles at rest. Clinically, normal muscle tone is identified as the slight resistance or stiffness that is felt in the muscles when the limb is moved passively at a slow, consistent rate of movement (between 2° and 12° per second) [40]. Normal muscle tone enables effective stabilisation of joints by balancing the tension in opposing muscle groups around a joint. Normal postural muscle tone assists in maintaining body posture with minimal energy costs [8, 38].

Historically, normal resting muscle tone has been described as the result of a constant, small amount of muscle contraction (due to the stretch reflex). However, no electromyographic (EMG) activity is typically found in normal muscles at rest or during slow passive stretch (the muscles are EMG-silent). Normal muscle tone, therefore, appears to be due to the intrinsic, passive rheologic (resistive) properties of the soft tissues. as described in Section 2.1.2.3 and Box 2.3. That is, the resistance or stiffness felt when ranging a relaxed limb that is not affected by brain injury is due to characteristics such as the elasticity and viscoelasticity of the surrounding connective tissues and muscle fibres, and the actin and myosin cross-bridges formed in the resting muscle [40]. Simons and Mense [41] have therefore defined normal resting muscle tone as “elastic and/or viscoelastic stiffness in the absence of contractile activity (motor unit activity)” (p. 3).

2.4.1.1 Muscle weakness

Muscle weakness can vary following brain injury, from partial or mild loss of strength (paresis) to severe (plegia) or complete (paralysis or flaccidity) loss of strength [3]. Weakness is a reduced ability to voluntarily generate and sustain sufficient force (or torque) for effective movement or task performance [19]. The consequences of muscle weakness include slowness to develop force, increased fatigability, reduced endurance and an increased sense of effort [65]. Flaccidity occurs when most spinal reflex responses, including stretch reflexes, are depressed, leading to an extreme form of muscle weakness. Flaccidity is usually an immediate and transient response to brain injury, however, when it persists, it is correlated with poorer functional outcome [19]. The importance of muscle strength for upper limb function is indicated by reports that the severity of paresis in the first month following stroke is the strongest predictor of upper limb functional outcomes at six months [66–68], and that voluntary finger extension and shoulder abduction within five days of stroke predict upper limb functional recovery [69, 70]. Wrist and grip strength are described as indicators of upper limb function for children with hemiplegia [71]. Further, overall upper limb strength is reported to directly influence activity measures and, therefore indirectly, to influence functional or participatory outcomes for children with cerebral palsy [72].

Normal muscular force production depends on the number and type of motor units recruited in a contraction, as well as their firing rates (see Section 2.1.2.1). Various physiological contributors to muscle weakness have been proposed following brain injury (Table 2.6). After stroke, motor units are recruited at lower-force thresholds (that is, less effort is required to recruit more motor units) and in a changed pattern. They also fire less rapidly and do not increase their firing rates appropriately in response to movement needs [65, 73–75]. It is possible that the inability to regulate firing rates within optimal ranges leads to greater reliance on over-recruitment in order to control force [40] and to difficulty producing force at shortened muscle lengths [76]. Clinically, over-recruitment of motor units presents as muscle overactivity, and inclusion of more muscles and joints than necessary for particular movements [77]. Difficulty in maintaining a constant force level due to reduced firing rates of agonist motor units may reduce performance of finely controlled movements and fine motor dexterity [18]. In contrast, prolonged firing rates may adversely affect the mechanical properties of associated muscle fibres, leading to fatigue and weakness, both post-stroke and in children with cerebral palsy [19, 78].

Selective loss of large motor units due to lack of stimulation (because of degeneration of the descending corticospinal tract at the segmental, spinal cord level) has been identified as an important contributor to muscle weakness [82]. Research has indicated that functioning motor units are halved in the first 6 months following stroke [83, 84] and that reductions are evident as early as 4–30 hours after stroke [85]. In addition, degeneration of type 2 (fast-contracting) motor units combined with an increased number or density of type 1 (slow-contracting) motor units reduces the force-producing capacity of adults with stroke and children with cerebral palsy [18, 65, 82]. Increased fibre density in muscles is an indicator of motor unit reorganisation in response to denervation of the muscle after motor neuron lesion. It develops because intact type 1 motor units send collateral nerve sprouts to previously denervated muscle fibres, leading to an increase in the number (density) of muscle fibres per motor unit in affected areas [82]. Post-stroke, the presence of a fourth type of motor unit characterised by slow contraction and increased fatigability is also reported to have been identified in people with long-standing hemiparesis [18].

In the upper limb, extensor muscles have been reported to be more affected by weakness than flexor muscles [76]. Reduced activity has been noted in shoulder abductor muscles although, in general, distal muscles on the impaired side have been identified as more affected, with greatest weakness found in wrist and finger muscles [86]. Evidence suggests that adults with hemiplegia due to stroke experience impairment of the ipsilateral (non-affected) upper limb in addition to the expected weakness in the contralateral (affected) upper limb [86, 87]. Similarly, children with either hemiplegic or diplegic cerebral palsy have impairments in their apparently non-affected upper limbs in addition to more severe impairment in their affected upper limbs [88]. These outcomes for the less-affected upper limb are proposed to reflect uncrossed fibres in the lateral corticospinal tract for both adults with stroke [89] and children with cerebral palsy [88].

2.4.1.2 Loss of selective muscle control

The severity of the central lesion experienced by a person with brain injury is thought to be reflected in his or her ability to isolate a movement without activating other parts of the limb, with more severe injuries leading to greater impairment in the person’s ability to coordinate limb segments for organised movement [90]. Following brain injury, loss of selective muscle control leads to difficulties with fractionation (isolation or differentiation) of movements, meaning that it is difficult to separate movement in one joint from movement in another joint in a limb. Thus, a person might consistently be observed to use the same inefficient and excessive movement pattern to try and achieve a number of movement goals [40, 42, 90]. For example, the person may habitually flex the elbow and abduct the shoulder while trying to reach forward, but may also use the same pattern when attempting to lift and carry an object.

In the UMNS, poorly differentiated muscle control may present in a variety of ways, including movement synergies, ataxia, dyspraxia or apraxia [40, 64, 75]. Each of these impairments adversely affects voluntary goal-directed movement [60], and this lack of upper limb control is often described functionally as reduced dexterity [11]. Ada and Canning [91] define loss of

“dexterity (motor control) … [as] a loss of coordination of voluntary muscle activity to meet environmental demands, and it is not restricted to manual dexterity … it is a loss of both the spatial and temporal accuracy needed [for] movement” [pp. 95, 97].

In addition to this definition, ‘dexterity’ has also been used to describe the capacity of isolating muscles for discrete movements. It is possible that disrupted transfer of sensorimotor information from the environment may affect selective muscle control and, therefore, both gross and fine motor upper limb dexterity [42].

The use of ‘obligate’ movement synergies or stereotypical patterns is one possible result of the inability to isolate relevant muscles after brain injury [40, 64]. A synergy is defined as “patterned movement of the entire affected limb in response to a stimulus or to voluntary effort” [92, p. 668], and is different to an associated reaction (see Section 2.4.2.8). Muscle synergies occur normally within the intact musculoskeletal system, their purpose being to reduce the degrees-of-freedom needing to be controlled to produce efficient limb use by providing a set of ‘ready solutions’ or ‘movement modules’ [93, 94]. However, in people with brain injury, movement synergies are consistently (inappropriately) used, regardless of the intended end goal of the movement. The precise reasons for such synergies are unclear. It has been suggested that brain lesions might interfere with the structure of the synergies that the CNS typically uses as motor modules for directing movement, therefore causing impaired action [93–95]. Another possibility is that post-injury plasticity promotes connections between the reticulospinal (brainstem) pathway and the corticospinal pathway. The reticulospinal pathway controls gross, multi-joint movement, while the corticospinal pathway controls distal, goal-directed forearm/hand movement. Connections between them may therefore lead to abnormal ‘neural coupling’ of the shoulder/arm and hand joints resulting in upper limb synergies [96]. Further, Roh et al. [95] have proposed that both these possible explanations for movement synergies may occur together, significantly reducing capacity for efficient, goal-directed movement. A number of upper limb synergies commonly observed in children with cerebral palsy and adults with stroke are presented in Table 2.7.

In children with cerebral palsy, ataxia is described as a negative feature of the UMNS. It is defined

“as an inability to generate a normal or expected voluntary movement trajectory that cannot be attributed to weakness or involuntary muscle activity about the affected joints” [64, p. 2162].

Thus, ataxia leads to impairment in joint positioning when moving the limb through space, as well as reduced accuracy due to an inability to activate the correct pattern of muscles for movement. Additional deficits considered to be components of ataxia include dysmetria (inaccurate movement leading to either under- or over-shooting an object), dyssynergia (degeneration of multijoint movements) and dysdiadochokinesia (difficulty performing rhythmic movements) [64]. In adults with stroke, ataxia is described as a loss of movement coordination, and is most apparent during fast, multi-joint movements in chronic stages of stroke [97].

Apraxia is another negative feature of the UMNS described as affecting the movement of children with cerebral palsy. It is defined as

“impairment of the ability to accomplish previously learned and performed complex motor actions that is not explained by ataxia, reduced selective motor control, weakness, or involuntary motor activity” [64, p. 2163].

Apraxia is distinguished by its presence during task-related activity, while ataxia is apparent during simpler, non-purposeful movements.

2.4.1.3 Learned non-use

Although not specifically described as a negative features of the UMNS, learned non-use of the hemiplegic upper limb following brain injury has become recognised as a significant contributor to decreased upper limb function [62]. Reduced use of the affected upper limb is proposed as being ‘learned’ as a response to the increased effort needed to move the limb due to muscle weakness combined with the inability to selectively control the amount and timing of movement needed to achieve intended tasks. Since movement is so effortful and functional outcomes so poor, people with hemiplegia are proposed as being less motivated to use the affected arm, resulting in learned non-use of that arm and hand [104]. In effect, an ongoing cycle of paresis/weakness-disuse-paresis/weakness is set up which can lead to tissue shortening and the development of contracture [18].

More recently, factors in addition to learning and motivation have been suggested as influencing non-use of the hemiplegic upper limb. Van der Lee [105] suggests that sensory disorders may be the most substantial contributor to learned non-use. This assertion is based on studies in which monkeys with deafferentation had little motor impairment but significant sensory impairment. In addition, Wolf [106] has used imaging studies with stroke-affected and healthy participants to propose that structural changes to areas of the brain involved in movement initiation (cingulate motor and pre-frontal areas) may cause reduced motivation to move prior to any learning effects.

2.4.2.1 Hyperactive tonic and phasic reflexes

Exaggerated, hyperactive or hyperexcitable spinal reflexes, evident either while the person is at rest or during movement, are responsible for most of the positive features of the UMNS that are observed clinically (Table 2.8). Spinal reflexes become hyperactive due to disruption of the regulating influences of the descending pathways involved in motor control, that is, the closely-associated corticospinal (pyramidal) and brainstem (parapyramidal) pathways. These pathways control stretch, flexor and extensor reflexes by modulating (adjusting) signals to the motor neurons in the spinal cord on the basis of sensory feedback from the periphery [11, 77]. It is most likely that a reduced threshold is the reason that the motor neuron is hyperactive. A reduced threshold means that, functionally, a smaller stretch at a slower than usual velocity results in a larger that usual response in the tested muscle. Studies suggest that, following UMN injury, motor neurons are constantly close to their threshold for excitation and therefore require little further input in order to fire [11, 61].

The stretch (or myotatic) reflex is the simplest example of a monosynaptic, short-latency, segmental spinal reflex. It comprises a receptor (muscle spindle), a sensory (Ia primary or afferent) neuron which takes information to the spinal cord, and a motor (alpha [α] or efferent) neuron that activates the fibres in the same muscle as the receptor (Figure 2.4). The stretch reflex is a proprioceptive reflex that is present in all muscles, both flexor and extensor, and is described as either phasic or tonic. It is triggered mechanically, by a rapid (phasic) stretch of the muscle, usually by moving the joint quickly through its range of motion or by tapping the relevant tendon. The stretch reflex that occurs following a tendon tap is known as a deep tendon reflex (DTR), a tendon reflex or a T-reflex. The tonic stretch reflex arises from a sustained muscle stretch and is the cause of spasticity (further discussed in Section 2.4.2.6) [33, 111]. See Box 2.4 for further information about reflexes.