Using the latency correction method to calculate the tonic stretch reflex threshold in spastic cerebral palsy pre and post exercise intervention: a case study

USING THE LATENCY CORRECTION METHOD TO CALCULATE THE TONIC STRETCH REFLEX IN SPASTIC CEREBRAL PALSY PRE AND POST EXERCISE INTERVENTION: A CASE STUDY Justin Frick Master`s Thesis Faculty of Sport and Health Sciences University of Jyväskylä Autumn 2024 ABSTRACT Frick, J. 2024. Using the latency correction method to calculate the tonic stretch reflex threshold in spastic cerebral palsy pre and post exercise intervention: a case study, Faculty of Sport and Health Sciences, University of Jyväskylä, Master’s thesis, 75 pp. 0 appendices. ABSTRACT Cerebral palsy (CP) is the most common childhood disability, caused by brain lesions during early development, often leading to motor-related symptoms and spasticity. Exercise interventions focusing on strength, gait, flexibility, and balance are valuable for managing CP. Tailoring exercise programs to individual needs may maximise outcomes. This case study was a smaller project related to the EXECP project. The case study aimed to evaluate changes in the tonic stretch reflex threshold (TSRT) following a 12-week exercise intervention for stretch hyper-reflexia in spastic cerebral palsy (SCP), aiming to provide an insight into the potential benefits of exercise on the TSRT in SCP. One participant with spastic cerebral palsy was selected (N=1) for the case study and performed passive muscle stretching of the soleus and medial gastrocnemius using a motor-driven dynamometer (Neuromuscular Research Center, University of Jyväskylä, Finland). The dynamometer induced passive ankle dorsiflexion from 20° of plantarflexion to 0° at four angular velocities (55, 110, 210, and 291°/s) as part of the EXECP study. The EMG data was then provided for this case study, allowing the TSRT to be calculated and assessed at four time points (PRE1, PRE2, POST1, POST2) and four velocities (55°/s, 110°/s, 200°/s, 291°/s) for the medial gastrocnemius and soleus. This case study used the EMG burst activity data from the EXECP intervention, and TSRTcorrected was calculated in Spike2 from Cambridge Electronics Design, Cambridge, UK . An individualized Hoffmanreflex (H-reflex) latency correction method was used, based on suggestions from Valadão et al. (2022) The h-reflex EMG data also came from the EXECP intervention. The case study analysed a portion of the h-reflex EMG data using MATLAB (R2023a, The MathWorks Inc, Natick, United States). The primary research questions are 1) is the TSRT corrected velocity dependent in this case study subject for the soleus and medial gastrocnemius? 2) how does TSRT corrected for the soleus and medial gastrocnemius muscles vary across different stretch velocities (55, 110, 200, and 291°/s) and 3) pre vs post exercise intervention. The soleus showed higher variability and sensitivity to velocity changes compared to the medial gastrocnemius, suggesting its role in adaptive postural control. After exercise, reflex thresholds in the soleus decreased and variability was reduced, indicating modulation of reflex activation. In contrast, the medial gastrocnemius showed stable, higher reflex thresholds, indicating its role in highspeed dynamic movements. The study reveals the distinct behaviours of the soleus and medial gastrocnemius, highlighting their specialised functional roles and neuromechanical properties. These findings suggest that the soleus is specialized for postural stability, while the medial gastrocnemius is suited for dynamic movements, informing targeted rehabilitation approaches. The medial gastrocnemius shows stable reflex thresholds, suited for dynamic, high-speed tasks, while the soleus exhibits more variability, reflecting its role in postural stability and controlled movements. These findings highlight the distinct roles of these muscles and suggest targeted interventions could enhance motor control in individuals with cerebral palsy, especially those with spasticity. The results must be interpreted cautiously, and no generalisations should be made as this was a case study. Key words: cerebral palsy, hyper-resistance, stretch reflex threshold, stretch hyper-reflexia, assessment, upper motor neuron syndrome, neuromuscular function. ABBREVIATIONS CPcerebral palsy MDDmotor development disorder SCPspastic cerebral palsy SRstretch reflex SRTstretch reflex threshold TSRT - tonic stretch reflex threshold TSRTcorrected - latency corrected tonic stretch reflex threshold MTLmotor tract lesion CNScentral nervous system PNSperipheral nervous system MSmultiple sclerosis iSCIincomplete spinal cord injury SCIspinal cord injury EMGelectromyography RMSroot mean square LMNslower motor neurons UMNSupper motor neuron syndrome UMNupper motor neurons GMFCSgross motor function classification system H-reflexHoffman reflex TDtypically developing individuals PCSAphysiological cross-sectional area MMSMontreal spasticity measurement SOLsoleus MGmedial gastrocnemius CONTENTS TIIVISTELMÄ ABSTRACT 1 INTRODUCTION ........................................................................................................... 2 2 LITERATURE REVIEW ................................................................................................ 5 2.1 Lesion mechanisms and pathophysiology of cerebral palsy ................................... 5 2.2 Cellular mechanisms of the brain injury and lesion ................................................ 8 2.3 Classification systems: functional, motor and dispersal models ........................... 10 2.4 Motor control systems ........................................................................................... 15 2.5 The stretch reflex and the stretch reflex threshold ................................................ 18 2.6 Neuromuscular and motor control alterations in cerebral palsy ............................ 20 2.7 Skeletal muscle morphology and function in cerebral palsy ................................. 23 2.8 Sarcomere structure and dynamics in cerebral palsy............................................. 24 2.9 Non-neural components of hyper-resistance ......................................................... 26 2.10 Exercise rehabilitation for cerebral palsy .............................................................. 28 2.11 The effects of exercise on the neuromuscular system ........................................... 32 2.12 Assessing stretch hyper-reflexia and the tonic stretch reflex ................................ 33 2.13 The passive muscle stretch technique .................................................................... 34 2.14 Electromyography and Surface Electromyography ............................................... 35 3 CASE STUDY AIMS, RESEARCH QUESTIONS AND HYPOTHESES ................. 37 3.1 Research questions and hypotheses ....................................................................... 39 4 METHODS .................................................................................................................... 40 4.1 Sampling and subject recruitment for EXECP ...................................................... 40 4.2 EXECP study design ............................................................................................. 40 4.3 Case Study design .................................................................................................. 43 4.4 Passive muscle stretch measurement protocol ...................................................... 45 4.5 EMG data collection and setup .............................................................................. 46 4.6 H-reflex recruitment protocol and setup ................................................................ 47 4.7 H-reflex latency calculation protocol .................................................................... 47 4.8 Data analysis preparation....................................................................................... 48 4.9 Filtering of data in spike ........................................................................................ 48 4.10 The tonic stretch reflex threshold corrected analysis ............................................ 49 4.11 Statistical analysis ................................................................................................. 50 5 RESULTS ...................................................................................................................... 51 5.1 Pre-intervention results .......................................................................................... 51 5.2 Post-intervention results ........................................................................................ 52 5.3 Summary of results across all time points ............................................................. 55 6 DISCUSSION ................................................................................................................ 57 6.1 Potential effects of the exercise intervention ......................................................... 59 6.2 Strengths and weaknesses of the study .................................................................. 63 7 CONCLUSION ............................................................................................................. 64 8 REFERENCES .............................................................................................................. 65 APPENDICES 1 INTRODUCTION Cerebral palsy (CP) is the most common childhood motor development disorder (MDD), caused by abnormal development or lasting non-progressive injury to the developing infant brain. It is a permanent, non-progressive condition affecting movement and posture (Graham et al., 2016a). Although classified as non-progressive, some symptoms may appear to worsen over time. CP arises during the prenatal, perinatal, or postnatal periods, typically resulting from a brain or motor lesion most often affecting the motor cortex, basal ganglia, or cerebellum (BarOn et al., 2018). Globally, CP occurs at rates ranging from 1.5 to 3 per 1,000 live births (Graham et al., 2016a; Johnson, 2002). Increased survival rates among premature infants may have slightly elevated CP prevalence, especially in developed nations. Preterm birth remains the most significant risk factor for CP, with risk decreasing as gestational age increases. The likelihood of CP rises moderately at 38 weeks, while infants born before 28 weeks face up to 50 times the risk of those born at full term (Kuban et al., 2009). Another major risk factor is white matter damage, which can be detected via ultrasonography or magnetic resonance imaging (MRI) (Graham et al., 2016b; Kuban et al., 2009). Features of CP is its clinical and functional presentation in early development, often involving persistent movement disorders. Motor impairments primarily result from damage to upper motor neurons, disrupting motor planning, execution, and feedback loop functions (Graham et al., 2016b). This frequently manifests as upper motor neuron syndrome (UMNS), with spasticity or stretch hyper-reflexia as key features. Spasticity is characterized by a velocitydependent increase in muscle tone due to hyperactive stretch reflexes, caused by imbalances in excitatory and inhibitory signals within the spinal cord and brain. These abnormalities exaggerate muscle responses to stretch, significantly impairing voluntary movement and functional abilities. Beyond motor impairments, CP is often associated with challenges in cognition, communication, sensation, perception, epilepsy, and musculoskeletal health (Graham et al., 2016a). Key clinical goals for individuals with CP include maximizing functional activities and participation, while minimizing the impact of associated conditions such as epilepsy, feeding difficulties, hip dislocation, and scoliosis (Graham et al., 2016b). Table 1 below gives a summary of cerebral palsy and its main pathologies and developmental changes. CP encompasses a wide range of complex aetiologies and pathophysiologies. Despite advances from epidemiologic, neuroimaging, animal model, and post-mortem studies, the mechanisms, features, and phenotypic variations of CP remain only partly understood. This limits comprehensive prevention strategies. Early clinical descriptions linked CP to premature birth and neonatal asphyxia, which were once thought to be definitive causes. However, recent research suggests these factors are reflective of broader developmental influences (Graham et al., 2016b). This case study aimed to examine the calculation of the stretch reflex threshold in an individual with spastic CP (SCP). It employed the stretch reflex threshold corrected method (TSRTcorrected) to calculate the stretch reflex threshold, as recently proposed by Valadão et al. (2022). The investigation focused on the TSRTcorrected in the soleus and gastrocnemius muscles of the triceps surae, which are critical for postural control and gait. Table 1: A summary of cerebral palsy (Lundy-Ekman & Weyer, 2022). Condition Cerebral palsy Pathology Developmental abnormalities Aetiology Abnormal development in the pre-, peri or post-natal period, metabolic anomalies, immune system syndromes, coagulation syndromes, infection, trauma, hypoxia, central nervous system damage before second birthday Speed of onset Unknown Signs and symptoms Consciousness Normal Cognition, language, and memory Often associated with intellectual disability, language challenges, although not all cases. Some individuals have above average IQ. Sensory 75% of cases present with pain, or impairments in somatosensation. Autonomic 25% of individuals experience incontinence. Motor 33% unable to walk, hypotonic individuals show low muscle tone, impaired movement capacity. Spastic type, paresis, muscle FIGURE 3. Cell death signalling pathway in brain injury related to CP from (Graham et al., 2016a). 2.3 Classification systems: functional, motor and dispersal models Cerebral palsy (CP) encompasses a group of permanent movement and posture disorders caused by non-progressive disturbances in the developing brain. Its classification is crucial for understanding its characteristics, planning management, and predicting outcomes. CP is categorized by motor type, impairment distribution, and severity, providing a framework to describe its diverse manifestations and guide therapeutic approaches (Graham et al., 2016a). Functional capacity classifications, discussed in the following sections, delve into the types, classifications, and impairment patterns across CP populations (Lundy-Ekman & Weyer, 2022). Since CP presents uniquely in everyone, classification systems consider functional impairments, muscle tone abnormalities, and impairment distribution to ensure tailored assessment and intervention (Graham et al., 2016a; Lundy-Ekman & Weyer, 2022). Motor type or muscle tone is based on the nature of the movement abnormality, with spastic, dyskinetic, ataxic, and mixed types being the primary categories. Spastic CP, the most common type, is characterized by increased muscle tone and stiffness, often leading to joint contractures. Dyskinetic CP involves involuntary movements, while ataxic cp is marked by difficulties with balance and coordination. Mixed CP presents a combination of these features (Graham et al., 2016b; Lundy-Ekman & Weyer, 2022). Dispersal classification focuses on the areas of the body affected. Terms such as hemiplegia, diplegia, and quadriplegia describe whether one side, the lower limbs, or all four limbs are involved. This categorization helps in understanding the extent of physical disability and planning mobility aids or interventions (Graham et al., 2016b; LundyEkman & Weyer, 2022). Functional classification, often assessed using tools like the gross motor function classification system (GMFCS), evaluates the degree of functional impairment. This system ranks individuals from level I (minimal impairment) to level V (severe impairment), guiding expectations and therapeutic goals (Graham et al., 2016b; Lundy-Ekman & Weyer, 2022). By combining these classification dimensions, clinicians can comprehensively describe the condition and predict developmental trajectories. This nuanced understanding of cp highlights its complexity while emphasizing the importance of individualized care. Each classification mode will be expanded in the below sections starting with functional classification, then muscle tone and finally dispersal (Graham et al., 2016b, 2016a). Cerebral palsy is commonly classified using the gross motor function classification system (GMFCS, see figure 4below) (Graham et al., 2016a). The GMFCS presents five different levels of functional classification for individuals with cerebral palsy. Classification at Level I characterise the ability of a child to perform gross motor tasks including running and jumping, with challenges in speed, balance, and coordination of movement present (Graham et al., 2016a). Level II classification represents additional challenges with endurance capacity, balance on uneven surfaces, and lower capacity when performing gross motor tasks. Furthermore, it may require use of an aid when walking. Level III, walking in most indoor settings requires a mobility device, for longer distances a wheelchair may be required. Stairs may be climbed using handrailing under supervision. Level IV, most often requires physical assistance or powered mobility devices to move. Walking short distances at home with physical assistance and or a body support walker. Outdoors and in the community motor powered mobility devices are mandatory. Level V indicates low muscle control of head, trunk posture, and limb movement. Mobility occurs in a wheelchair in all situations (Graham et al., 2016a). Cerebral palsy may also be categorised based on the muscle tone observed, often represented as dystonia, ataxia, or spastic. Cerebral palsy is commonly presented in five different groups, with four of the five types showing abnormal muscle tone. Muscle tone is defined as the amount of resistance to stretch in a relaxed muscle, in a healthy neuromuscular system relaxed muscle provides little resistance to passive stretch (Lundy-Ekman & Weyer, 2022). However, in cerebral palsy muscle tone is abnormal and muscle tone is presented on a scale from hypotonia to hypertonia. Hypotonia representing lower than normal resistance, while hypertonia, represents greater than normal resistance to stretch. The classification of abnormal muscle tones represents the muscle tone present (Graham et al., 2016a; Lundy-Ekman & Weyer, 2022). Hypotonic CP, is categorised due to very low muscle tone, associated with inadequate muscle contraction required to maintain regular head and trunk postures. Individuals with hypotonic CP have little to no capacity to actively move. Spastic CP, and spasticity represents the velocity dependant component of hypertonia. This type is considered velocity dependent as there is little resistance at low stretch velocities, with increasing resistance present at higher stretch velocities. This makes muscle more stiff than normal often leading to contractures, due to abnormally short muscles and tendons. Often leading to abnormal patterns in the gait cycle, seen in toe walking or scissor gait (Meyns et al., 2016). Furthermore, spastic CP is categorized based on the area of the body that is impacted, termed hemiplegic, tetraplegic, or diplegic. These terms will be explained in the following section. Dyskinetic CP is a term used to describe fluctuating states of muscle tone. Most seen is choreoathetoid with dystonia. Represented by three types of spontaneous movements, choreiform, athetoid or dystonic. Choreiform includes jerky, abrupt, or irregular movements, while athetoid presents as slow, or writhing movements and dystonic represents with spontaneous sustained muscle contraction. In addition, mixed type CP represents a combination of spasticity and dyskinesia. The final classification given to CP is ataxic, this type does not present with abnormal muscle tone, instead movements are uncoordinated and shaky during voluntary movement (Graham et al., 2016b; Lundy-Ekman & Weyer, 2022). FIGURE 4. The Gross motor function classification systems (GMFCS) for children with cerebral palsy from (Graham et al., 2016a). Classification is often presented in combination with an impairment dispersal, referred to as either unilateral or bilateral, see figure 5 below (Bar-On et al., 2018). The two impairment dispersals can further be divided, unilateral may represent monoplegia or hemiplegia. Monoplegia usually affects one limb, most frequently the lower limb. Hemiplegia usually affects one side of the body, with the upper limbs most frequently affected. While bilateral impairment presents three groups diplegia, triplegia or quadriplegia (see figure 5 below). In diplegia each limb is affected, with lower limbs more than upper limbs. Triplegia, represents unilateral upper limb impairment, combined with bilateral lower limb impairment. The lower limb is most frequently affected on the same side as the upper limb impairment. Quadriplegic involvement represents all limbs and the trunk, it may also be referred to as whole body involvement. Therefore, the levels of impairment faced by individuals with cerebral palsy are highly individualised, based on the severity and complexity of the lesion (Graham et al., 2016b; Sloot et al., 2021). FIGURE 5. Unilateral and bilateral cerebral palsy description taken from (Graham et al., 2016a). 2.4 Motor control systems A motor control system is comprised of the following components: the cerebral cortex, responsible for processing and commanding motor signals for performance. The frontal motor area, pre and supplementary motor areas, prefrontal cortex, parietal cortices, and association areas responsible for conscious and unconscious visual, auditory, tactile, and proprioceptive feedback (Mukherjee & Chakravarty, 2010). Combined with subcortical areas of the basal ganglia, cerebellum, and brainstem. Notably the red nuclei are involved in stretch reflexes down to the spinal cord which consists of motor units and spinal cord reflexes. Including the cutaneous reflex, withdrawal reflex and muscle reflex which is of major interest in the case of stretch hyper-reflexia (Lundy-Ekman & Weyer, 2022). The most important motor functions related to stretch hyper-reflexia are related to muscle receptors and muscle stretch reflexes. In this case the tonic (static) stretch reflex by passive stretching is used to test muscle tone and excitability in the case of stretch hyper-reflexia (Lundy-Ekman & Weyer, 2022; Mukherjee & Chakravarty, 2010). Muscle function is largely dependent on the excitement of anterior horn motor neurons, combined with sensory feedback from muscle spindles to the spinal cord. Muscle spindles are responsible for giving feedback on the length and tension of a muscle (Lundy-Ekman & Weyer, 2022). They consist of two types of fibres, intrafusal and extrafusal. Intrafusal fibres are innervated by smaller gamma efferent fibres, while extrafusal fibres are innervated by large alpha efferent fibres. They are also often separated into two types of sensory endings, primary and secondary. Primary receptors are group Ia afferent fibres and secondary are group II afferent fibres (Enoka, 2015). The myotatic reflex, is a role carried out by the muscle spindle. When a muscle is stretched this causes excitation of the spindle and causes the reflex contraction of the muscle and its synergists. In contrast the dynamic stretch reflex is elicited when a fast stretch of the muscle is performed (Lundy-Ekman & Weyer, 2022). The result is stimulation of Ia afferent fibres from the nuclear bag of the muscle spindle through the monosynaptic pathway. The dynamic response subsides rapidly within milliseconds if a sustained static stretch continues. The static stretch is facilitated by the nuclear chain fibres of the muscle spindle, via group II afferents and certain Ia afferent fibres, which act via interneurons polysynaptically (Mukherjee & Chakravarty, 2010). Muscle tone is maintained by muscle spindles acting through the stretch reflex. See figure 6 below for reference on muscle spindle structure. This background activation is necessary to maintain posture and stability. Tone is responsible for keeping muscle at a given length and is altered during movement to maintain stability. Gamma fibres are mostly responsible for this action and react to any signals sent to alpha motor fibres (Enoka, 2015). This is termed alphagamma co-activation and is present during contraction, to contract both intra and extrafusal fibres dependent on the position and force output from the brain towards the spinal cord. The stretch reflex is often tested in clinical settings using one of two tests. Static testing is performed by passive stretching or tone testing. While dynamic testing occurs via muscle and tendon jerk techniques (Lundy-Ekman & Weyer, 2022; Mukherjee & Chakravarty, 2010). FIGURE 6. Muscle spindle outlining key structural components from (Mukherjee & Chakravarty, 2010) Furthermore, interneurons play a vital role in mediation of spinal cord function, due to excitation or inhibition. As they are implicated in all segmented and stretch reflex pathways, via peripheral and descending fibre tracts. Interneurons involved in spasticity will be examined in the following sections. Renshaw cells located in the ventral horn medially to motoneurons in lamina VII (Enoka, 2015; Mukherjee & Chakravarty, 2010). Signals from alpha motoneuron axons also stimulate Renshaw cells, in response inhibiting these motoneurons and others innervating the synergist muscle group/s. This creates a negative feedback loop along the alpha motor neuronRenshaw cellalpha motor neuron pathway. Resulting in recurrent inhibition, ultimately altering motor neuron excitability. Additionally, Renshaw cells appear to inhibit gamma motor neurons and Ia inhibitory interneurons (Mukherjee & Chakravarty, 2010). Another vital role of interneurons is reciprocal Ia inhibition. When a muscle is stretched, Ia afferent fibres are activated producing monosynaptic excitation of homonymic alpha motor neurons. Combined with disynaptic inhibition of alpha motor neurons of antagonistic muscle groups, thus resulting in reciprocal inhibition (Enoka, 2015; Lundy-Ekman & Weyer, 2022). Ia interneurons obtain the same excitatory and inhibitory inputs from segmental afferents and supraspinal tracts as alpha motor neurons. Reciprocal inhibition reduces activity in antagonist when the agonist is active, thus the agonist can act unrestricted. This prevents unwanted antagonist activity when an agonist is voluntarily activated. Allowing for optimal output of the agonist muscle groups as they are unopposed by their antagonist counterparts (Castle et al., 1979; Mukherjee & Chakravarty, 2010). Interneurons are also responsible for inhibiting group II afferents, combined with their role in the stretch reflex arc. They are also responsible for production of the flexor extensor response via excitation of flexor alpha motor neurons and inhibition of extensor motor neurons. Moreover, interneurons are responsible for non-reciprocal inhibition, presynaptic inhibition, and flexor reflex afferent activity (Lundy-Ekman & Weyer, 2022). Presynaptic inhibition is related to the amplitude of electrical post synaptic potential (EPSP), as a response to Ia afferent stimulation which is affected by preceding depolarization of the Ia afferent fibre via axo-axonic synapses within exact interneurons. The exact interneurons responsible for presynaptic inhibition are commanded via descending tracts, resulting in autonomous inhibition of insignificant afferent outputs (Enoka, 2015; LundyEkman & Weyer, 2022) . The mechanisms related to spasticity are complex, however relatively well understood. It appears that excitatory and inhibitory mechanisms both play a role in spasticity. To what extent is still unclear, however it is hypothesised that inhibitory processes play a greater role than excitatory processes (Mukherjee & Chakravarty, 2010). Correlated to the pathophysiology of spasticity and spastic paretic syndrome, the mechanisms are broadly grouped into two categories. Firstly, spinal mechanisms which are related to the spinal neurons and motor systems. Secondly, they may be grouped into supraspinal and suprasegmental mechanisms. Prior to expansion of these mechanisms, it is worth mentioning the components of both motor control systems and the motor functions of the spinal cord (Mukherjee & Chakravarty, 2010). Spinal reflexes play a vital role in human movement and motor control. A reflex is a nonvoluntary movement provoked by stimulation of sensory receptors. In the case of stretch hyperreflexia the reflex response is altered due to damage to the motor tract, resulting in hyperactivity of the reflex arc and disruption of muscular function. This results in disinhibition of the alpha motor neurons. 2.5 The stretch reflex and the stretch reflex threshold The stretch reflex (SR) is a fundamental neuromuscular response that occurs when a muscle is stretched, triggering an automatic contraction to counteract the stretch and maintain stability. It plays a crucial role in maintaining muscle tone, posture, and joint integrity (Lundy-Ekman & Weyer, 2022). The SR involves three key components: muscle spindles, Ia afferent neurons, and alpha motor neurons. Muscle spindles, located within the muscle fibres, act as sensory receptors that detect changes in muscle length and the speed of stretch (Enoka, 2015). When a muscle is lengthened, these spindles generate signals transmitted by Ia afferent neurons to the spinal cord. There, the signal activates alpha motor neurons, which initiate a reflex contraction in the stretched muscle (Enoka, 2015). The stretch reflex threshold (SRT) refers to the specific muscle length or joint angle at which the stretch reflex is initiated. It is modulated by descending signals from the central nervous system (CNS), allowing the body to adapt muscle tone based on functional demands. For example, during walking, SRT regulation ensures fluid, efficient movement by preventing premature muscle activation (Enoka, 2015). However, in neurological disorders like CP, this modulation is disrupted. In typically developing individuals, the SRT is modulated by descending neural input from the brain, which adjusts muscle tone to meet functional demands. However, in individuals with CP, damage to UMNs disrupts this modulation, lowering the SRT. As a result, stretch reflexes are triggered prematurely, leading to excessive muscle activation, stiffness, and impaired movement control (Lundy-Ekman & Weyer, 2022). In CP, pathophysiological changes to the SRT result from upper motor neuron (UMN) lesions, which impair the balance between excitatory and inhibitory signals in the spinal cord. In CP, the neurological damage to the UMN system significantly alters the regulation of the SRT. Normally, the UMN system provides descending inhibitory signals that suppress the overactivity of reflex pathways, maintaining an optimal SRT. However, damage to the motor cortex, basal ganglia, or corticospinal tracts disrupts these pathways, leading to disinhibition of spinal reflexes and an abnormally low SRT often seen in CP (Valadão et al., 2022). Consequently, stretch reflexes are triggered more readily and at inappropriate muscle lengths, resulting in spasticity and involuntary co-contractions (Levin et al., 2024). A lowered SRT triggers reflex activation prematurely, resulting in hyperreflexia, an exaggerated reflex response. This contributes to spasticity, a condition characterized by velocity-dependent increases in muscle tone. The combination of hyperreflexia and spasticity disrupts voluntary movement control, leading to stiffness, reduced range of motion, and poor coordination. This altered SRT profoundly affects motor function, making it a key target for rehabilitation strategies (Valadão et al., 2021). The SRT is critical for maintaining an optimal balance between muscle stiffness and flexibility. In CP, a reduced SRT contributes to spasticity, a condition defined as hyperactive muscle responses to stretch. This hyperreflexia leads to, difficulty in controlling voluntary movements, limited joint range of motion due to persistent muscle activation, abnormal gait patterns, such as scissoring or toe-walking. This disrupts gait efficiency and increases the energy cost of movement (Meyns et al., 2016). Additionally, spasticity and hyperreflexia restrict joint mobility, making tasks like sitting, standing, or transitioning between postures more difficult (Graham et al., 2016b). Furthermore, functional limitations associated with an altered SRT often lead to secondary complications, such as joint contractures, deformities, and chronic pain. Over time, these issues exacerbate the physical challenges faced by individuals with CP. Addressing the SRT through targeted rehabilitation strategies could reduce spasticity and improve motor control, enabling better movement efficiency and functional independence. Given the centrality of the SRT to motor impairments little. Hypothetically, the number of cross bridge cycles is altered in person with CP, that is fewer cross bridges are able to cycle, resulting in lower-than-expected force production (Lieber & Fridén, 2019; Smith et al., 2011). As the number of cross bridges in series, allows for alterations in force production capacity. That is the more cross bridges in series, the greater potential for force production (Lieber & Fridén, 2019). This must be considered in combination with factors such as tendon stiffness, conduction velocity of nerves, and voluntary activation which also affect force production capabilities. 2.9 Non-neural components of hyper-resistance Non-neural components are related to the muscle tissue properties, identified via three components elasticity, viscosity and shortening properties. The non-neurological component alterations are believed to occur because of muscular adaptations caused by neurological dysregulation. This is seen when muscle contractures occur due to shortened muscles or stiffen due to architectural changes in the muscle tissue. These maladaptive changes may be magnified in children due to the growth process and its effect on tissue development and motor control (Graham et al., 2016a). An explanation may be due to the alterations in musculo-tendinous properties in CP due to chronic changes, caused by contractures. The changes at the muscle and tendon level, may be induced as a biproduct of altered motor control or due to physiological processes which may be altered in CP, due to the motor lesion itself (Sloot et al., 2021). Altered stretch reflex activity has been associated with gait pattern alterations seen in CP. Caused by limited ankle dorsiflexion during the contact, swing, and stance phases of the gait cycle, combined with decreased push of power at the terminal phase (Meyns et al., 2016). The hyperactivity is often related to chronic maladaptation in the musculoskeletal system, seen via contractures, increased muscle stiffness and deformation in developing skeletal bones (Peeters et al., 2023). Due to this assumption aggressive treatment methods are commonplace with the aim to decrease or postpone orthopaedic surgery. Other studies have also argued that hyperreflexia plays and insignificant role in gait changes (Nielsen et al., 2005). Thus, it is important that we can assess muscle-tendon interactions, during passive and active movements. Previous studies have suggested that the medial gastrocnemius (MG) muscle belly has a reduced capacity for lengthening during low-speed passive ankle rotations (Barber et al., 2012; Kalkman et al., 2018). Additionally, during the mid-stance phase of walking muscle fascicles appear to lengthen more, thus suggesting altered musculo-tendinous interactions (Kalsi et al., 2016). These results suggest that muscle-tendon unit function is altered in both passive and active states. The resultant outcome appears to be an increase in fascicle lengthening, which may alter muscle spindle activity (Bar-On et al., 2018). Leading to increased activation of muscle spindles, and increased Ia afferent reflex arc activity, resulting in increased SR response (hyper-activity). Furthermore, (Graham et al., 2016a) hypothesised that muscle architectural changes may be altered in children with cerebral palsy (see figure 6). This may be attributed to differences in sarcomere length, although similar fascicle lengths may appear. Graham et al., 2016 suggested this may be due to the way that sarcomeres are added in series during bone maturation (growth), suggesting the sarcomeres may not be added in series in individuals with cerebral palsy. Therefore, when the bone grows the ankle is pushed into plantarflexion due to structural maladaptation, and eventually contractures. (Graham et al., 2016a). In addition, suggestions of increased extracellular matrix (ECM) may alter MTU interaction, further helping to explain increased muscle spindle activity. A summary of factors contributing to muscle resistance to stretch has been adapted from (Lundy-Ekman & Weyer, 2022). The main factors are summarised in the figure 9 below are, descending motor commands, proprioceptive information, connective tissue properties, weak cross-bridge binding, and titin, all forming a muscles resistance to stretch. FIGURE 9. Summary of factors affecting muscle resistance to stretch created with information from (Lundy-Ekman & Weyer, 2022) 2.10 Exercise rehabilitation for cerebral palsy Unfortunately, there is currently no cure for CP, however exercise and physical activity (PA) have the potential to impact health outcomes and activities of daily living. Individuals with cerebral palsy most often have reduced physical activity resulting in lower cardiorespiratory fitness and muscle strength. (Verschuren et al., 2016), presented the figure 10 attached below as an example of the changes in PA seen due to GMFCS. It clearly identifies the change in PA due to GMFCS scores, with higher scores correlating with lower scores in PA. This is logical as we would expect lower scores from less mobile individuals with significant alterations in motor control. Based on a review of the current literature, it appears that individuals with CP spend between 76% to 99% of their days being sedentary, less than 18% partake in light PA, and between 2% to 7% perform vigorous exercise (Reid et al., 2015). Thus, the potential for development of chronic health conditions is increased. Poor physical activity has been studied extensively, it may lead to the development of chronic metabolic and cardiovascular disease. These may include diabetes, asthma, hypertension, stroke, joint pain and arthritis. Therefore, improving the amount of physical activity and health outcomes presents a vital part of the Muscle resistance to stretch Descending motor outputs Prioprioceptive information and feedback Connective tissue properties Weak cross-bridge binding Titin exercise paradigm for individuals living with cerebral palsy (Verschuren et al., 2016). The recent guidelines set out by (Verschuren et al., 2016), may provide us with a starting point for physical activity guidelines for those with CP see table 2 below. Exercise rehabilitation presents an important and unique tool for exercise, health and medical professionals in managing and treating cerebral palsy, not just to manage the disease but prevent further potential for the development of comorbidities. Recent evidence suggests that the timing of rehabilitation may affect the response and outcomes of the intervention, with earlier interventions appearing to offer the greatest potential (Reid et al., 2017; Verschuren et al., 2016). The benefits and use of exercise for cerebral palsy rehabilitation are relatively well studied, however optimal use and prescription of exercise remains unelucidated (Valadão et al., 2021). Exercise interventions most often use strength training, gait training, flexibility training or a combination of several methods. Commonly used methods include constraint induced movement therapy (CIMT), bimanual intensive therapy (BIT), tailored exercise interventions, gait training and balance training (Lundy-Ekman & Weyer, 2022). Exercise prescription for individuals with CP should consider a few important factors, the type, subtype, and brain lesion size and locality. Furthermore, the age and cognitive function of the individual should always be considered in developing exercise interventions, as this may affect their implementation and safety (Reid et al., 2015). Other factors which should be considered as with any exercise programs are, the mode, dose, context, relevance and timing of the program (Verschuren et al., 2016). Mode relates to the type of exercise performed, and its relevance to the motor qualities being trained. Dose is the amount of exercise performed, it combines both duration and intensity of exercise. By altering the dose the outcomes may change, thus finding a point suitable for everyone is of upmost importance. Relevance refers to the relevance of the exercise type in helping improve relevant outcomes. This will likely vary depending on the individual, hence our study for individualised exercise programs (Verschuren et al., 2016). The main aim of exercise interventions is to cause adaptation in the relevant systems. In this case the aim is to alter nervous system, and muscle structure and function, which effects the function of the neuromuscular system (Reid et al., 2015). The neuromuscular system is highly plastic in nature and responds well to training in TD individuals. However, so far data is mixed in identifying the outcomes and changes in the neuromuscular system in CP (Moreau & Lieber, 2022). Consideration of the individualised responses to exercise as a unique biological entity combined with a disordered state. Thus, individualised programming may help maximise relevant outcomes dependent on the individual's needs and condition. Another important consideration is the amount of exercise that is functionally possible for each individual based on their functional capacity (Verschuren et al., 2016). A key challenge in sustaining physical activity is lifestyle change, which can be especially difficult for people with CP due to various barriers. These include physical limitations, environmental factors, and caregiving challenges, such as time constraints, financial strain, and emotional stress. Families and caregivers have identified these barriers as significant obstacles to meeting physical activity guidelines (Verschuren et al., 2016). Given these challenges, it is crucial to identify individuals who would benefit from exercise interventions to mitigate long-term health risks. One of the most effective tools for this is cardiopulmonary exercise testing, which is the gold standard for assessing exercise tolerance and cardiorespiratory endurance. Exercise testing can help clinicians identify those at risk for poor health outcomes and determine who would benefit from targeted interventions. Despite its clear relevance, clinical exercise testing is underused, primarily due to a lack of clinician understanding and training on test administration and interpretation (Verschuren et al., 2016). To address these gaps, clinicians should encourage physical activity and consider recommending exercise testing when individuals with CP experience physical exhaustion or limitations in activity. The first step in designing exercise interventions is to assess physical fitness through objective testing. Once fitness levels are determined, it is important to identify the causes of deconditioning, whether due to inactivity, poor nutrition, disease-specific pathophysiology, or a combination of factors (Verschuren et al., 2016). There are established, clinically feasible exercise tests for individuals with CP, supported by solid evidence of their effectiveness. Much of the current research on physical activity benefits comes from studies involving ambulatory children and adolescents with CP. However, applying these findings to real-world interventions remains challenging due to practical limitations and a lack of inclusion of such applications in research (Moreau & Lieber, 2022). Figure 10. Physical activity based on GMFCS scores, SBsedentary behaviour, LPAlow physical activity, MVPAmoderate-vigorous physical activity from (Verschuren et al., 2016). To address this, it is vital to establish guidelines and protocols that promote physical activity as part of regular clinical and therapy centre visits for individuals with CP. These guidelines will help improve clinician comfort with discussing physical activity and guide future intervention studies to further promote healthier lifestyles and improve long-term health outcomes for this population. Table 2 below is comprised of exercise recommendations for cerebral palsy and gives a good outline of the type, frequency, intensity and time all key components of exercise programming (Verschuren et al., 2016). Table 2. Exercise recommendations for cerebral palsy created with information from (Verschuren et al., 2016). Exercise Frequency Intensity Time Type Cardiorespiratory fitness 1-2 sessions per week working towards three. >60% of HR max, or 40% of HRR, or between 46-90% of VO2peak. At least 20 minutes per session, for at least 8-16 weeks *frequency dependant. Regular purposeful exercise involving major muscle groups. Resistance training 2-4 times per week on nonconsecutive days. 1-3 sets of 6-15 repetitions. Equivalent to 5085% of 1RM No specific details mentioned, but duration of 12-16 weeks. Progressive based on single and multi-joint exercises. Daily physical activitymoderate to vigorous >5 days per week Moderate to vigorous intensity 60 minutes Varied modes and activities Daily physical activitysedentary 7 days per week Sedentary <1.5 METs <2 hours per day or break up sitting every 30-60mins. Non-occupational, leisure time activities, such as watching TV, using computer or playing games. 2.11 The effects of exercise on the neuromuscular system Exercise has a profound impact on the neuromuscular system, which includes both the muscles, and the nervous system involved in movement and coordination. Regular physical activity enhances neuromuscular function by promoting neural adaptations, such as improved motor unit recruitment and coordination between muscles and nerves (Enoka, 2015). This leads to more efficient movement patterns and an increase in muscle strength and endurance. Resistance training, for example, can stimulate the growth of muscle fibres and the recruitment of additional motor units, improving overall muscle performance and force production. Additionally, exercise has been shown to enhance neuromuscular plasticity, allowing the system to adapt to new movement demands and recover from injury more effectively (LundyEkman & Weyer, 2022). One of the key effects of exercise on the neuromuscular system is the improvement in muscle fibre composition and muscle function. With consistent strength training, there is an increase in the cross-sectional area of muscle fibres, particularly type II (fast-twitch) fibres, which contribute to greater force production (Enoka, 2015). In addition, endurance exercise improves the efficiency of type I (slow twitch) muscle fibres, which are critical for sustained activity and fatigue resistance. These changes result in better overall muscle performance, including increased strength, flexibility, and endurance. The neuromuscular system also adapts by increasing the number of motor neurons and enhancing the connection between the brain and muscles, which leads to improved motor control and coordination (Enoka, 2015; McArdle et al., 2010). Exercise also has a significant impact on the central nervous system (CNS), enhancing brain function and motor control. Regular physical activity has been linked to improved synaptic plasticity, which refers to the strengthening and formation of new connections between neurons (Lundy-Ekman & Weyer, 2022). This can lead to better motor learning, quicker reaction times, and enhanced motor planning and execution. For individuals recovering from neurological impairments or injuries, such as stroke or spinal cord injury, exercise promotes neuroplasticity, supporting the brain's ability to reorganize itself and regain lost functions. In particular, exercises that require complex movements or coordination can help retrain the CNS and facilitate recovery of lost motor functions (Lundy-Ekman & Weyer, 2022). However, the effects of exercise on the neuromuscular system can vary depending on the type, intensity, and duration of the exercise, as well as the individual's baseline health and neurological condition. In people with neurological disorders such as cerebral palsy or multiple sclerosis, the neuromuscular adaptations may be less pronounced, and exercise interventions need to be tailored to the individual’s specific needs and capabilities (Lundy-Ekman & Weyer, 2022). While regular exercise can improve neuromuscular function in these populations, careful consideration of the mode, intensity, and progression of exercise is essential to avoid overexertion or injury. With appropriate modifications and guidance, exercise can significantly enhance the neuromuscular system’s function and contribute to improved quality of life and mobility (Moreau & Lieber, 2022; Verschuren et al., 2016). 2.12 Assessing stretch hyper-reflexia and the tonic stretch reflex Assessing hyper-reflexia and hyper-resistance is crucial for identifying clinical status, planning treatment, and evaluating intervention success. These assessments are typically divided into subjective and objective measures. Subjective methods, such as the modified Ashworth or Tardieu scales, remain dominant in clinical practice due to their ease of use and practical implementation in a clinical setting, guiding decision-making and treatment planning (Graham et al., 2016b). On the other hand, objective measures, which involve manual or motorized instrument assessments, tend to be more reliable (Sloot et al., 2017), though they are often timeconsuming and require the operation of complex equipment. This raises the question of whether different methods assess similar characteristics, leading to comparable conclusions that inform health management and planning (Graham et al., 2016b). An important consideration is the time required to perform more objective tests in clinical settings. These tests are typically more complex to organize and conduct, often necessitating additional training for practitioners. While both subjective and objective assessments are crucial, the current methods often remain highly subjective (Malhotra et al., 2009), highlighting the need for reliable, repeatable, and valid objective measures. Another key issue in using the stretch reflex threshold (SRT) is the method of calculation. Valadao et al. (2022) suggest that much of the literature contains a systematic error in SRT identification due to the stretch reflex latency—the time between the mechanical initiation of the stretch reflex (SR onset) at the muscle spindle and the onset of the EMG signal (SR EMG onset). The variance between the joint angle at SR onset and SR EMG onset (SRT) is positively correlated with stretch velocity. As velocity increases, the error in joint angle calculation also increases due to the monosynaptic SR latency of approximately 30ms. This results in an overestimation of the SRT at higher velocities. Although Levin and Feldman (1994) proposed a latency correction method, later studies did not adopt this approach, introducing a systematic error in SRT calculation. SR latency primarily depends on individual anthropometrics and axonal pathways to the target muscles. Therefore, subtracting a fixed 30ms average latency could introduce a subject-specific error (Valadão et al., 2022). To address this, it is recommended to calculate and subtract the individual SR latency from the SRT. This would help assess the impact of the correction method on the linear relationship between SRT and stretch velocity and validate the TSRT method with latency correction (Valadão et al., 2022). Valadão et al. (2022) implemented latency correction by measuring Hoffman reflex (H-reflex) latency 2.13 The passive muscle stretch technique The passive muscle stretch technique, often performed using a motor-driven dynamometer, is commonly used to assess hyper-resistance and hyper-reflexia. This method allows for the simultaneous collection of kinematic, torque, EMG, and ultrasound data by performing passive movements at various velocities and accelerations to differentiate between neural and nonneural components of hyper-resistance. Fast movements help identify velocity-dependent neural components by detecting EMG bursts at the stretch reflex threshold (SRT), with higher velocities leading to stronger EMG bursts (Sloot et al., 2021). Acceleration also influences SRT, with higher peak accelerations leading to earlier EMG burst onset (Sloot et al., 2021). Slower movements are better suited for identifying involuntary background activation and non-neural tissue components (van den Noort et al., 2017), with non-velocity-dependent movements assessing EMG levels and the EMG-joint angle interaction, while non-neural components focus on torque-angle interactions. Figure 12 below represents an example of SRT identification as performed by (Bar-On et al., 2018). Typically, the subject is seated with knee and hip joints flexed, and the foot placed in the dynamometer’s footplate to measure torque and passive resistance. EMG data is often recorded from the Soleus (Sol), medial gastrocnemius (MG), and lateral gastrocnemius (LG) during passive plantar and dorsiflexion, following SENIAM guidelines (Hermens et al., 2000). The motor then performs a set number of movements at specified velocities or accelerations, often using randomized patterns. The resulting data includes range of motion, peak torque, passive torque, SRT, maximum angular velocity, average RMS EMG, work, and stiffness. According to van den Noort et al. (2017), slow movements produce variables such as ROM, maximum angular velocity, average RMS EMG, and SRT, while fast movements yield data on angular velocity, RMS EMG, SRT, work, angle of catch (AOC), and intensity of catch. While the passive muscle stretch technique is a reliable, reproducible, and valid method for research, its equipment can be bulky and restrictive. Additionally, it isolates the test limb, which may not fully represent real-life activities like gait (Sloot et al., 2017). Nonetheless, it provides valuable objective and quantitative data on hyperresistance, which can inform clinical decision-making for patient treatment (Graham et al., 2016b). 2.14 Electromyography and Surface Electromyography Electromyography is a technique which is used to identify the electrical activity generated in a muscle. In CP it helps us to identify the TSRT see figure 11 below for an example from (Sloot et al., 2017). The EMG signal represents a summation of the action potentials and is the product of nervous system activity and physiological activities of the muscle (Enoka, 2015). One challenge that may be encountered when using sEMG is the noise which may be seen due to objective and quantitative outcomes with minimal bias, despite potential measurement variability. FIGURE 13. Flow chart for the EXECP intervention outlining participant selection, testing and exercise intervention from (Valadão et al., 2021). 4.3 Case Study design This case study involved a single subject randomly selected for analysis, adhering to the EXECP sampling and recruitment protocol previously outlined above and by Valadão et al. (2021). This case study included a single participant with spastic cerebral palsy (N=1). To qualify, the participant had to complete all components of the EXECP exercise intervention and relevant testing session 1. The study focused on an individual with spastic cerebral palsy (N=1) and formed part of a broader investigation EXECP, as detailed above. This exploratory study is one of the first to examine TSRTcorrected longitudinally in this context. It utilized quantitative data collection methods, including electromyography (EMG), to provide a detailed analysis of the h-reflex latency and TSRTcorrected for the case study subject. Ethical approval was as part of the EXECP intervention and granted by the Central Finland Healthcare District Ethics Committee (U8/2017). The project adhered to the University of Jyväskylä’s data management policies. Participant confidentiality was maintained through anonymization of data, and informed consent was obtained from participants and/or their guardians. Data were securely stored and managed accordance with institutional standards. Testing sessions All testing sessions were conducted as part of the EXECP study. Testing sessions were conducted at the NMRC and the Centre for Interdisciplinary Brain Research (CIBR) at the University of Jyväskylä. Each testing phase (pre-test 1, pre-test 2, and post-test 1) comprised three separate sessions (S1, S2, and S3), scheduled 4 hours to seven days apart. Post-test 2, however, included only S1 and a modified version of S2. Session 1 (S1) included assessments of joint flexibility, muscle electrostimulation (H-reflex and PAD), muscle strength, and the tonic stretch reflex threshold (Valadão et al., 2021). Session 1 was the session relevant to the case study performed. The exercise training intervention was completed as part of the EXECP project and was finished prior to the commencement of this study. The intervention information is included as it is relevant to the current studies research questions. The exercise intervention consisted of strength training, gait training and flexibility training. The strength training program included five multi-joint and five single-joint exercises, designed to balance strength development and motor control while minimizing compensatory movements. Two alternating session designs (A and B) targeted the lower limbs and trunk with 7–10 exercises per session. Sessions alternated in AB or ABA/BAB patterns, depending on training frequency. Exercises for the ankle plantar flexors included seated and standing calf raises, while dorsiflexors were trained using resistance bands or manual resistance during hip extension exercises. Upper leg training incorporated leg press, squats, and knee flexion/extension, with unilateral exercises addressing limb imbalances. Trunk training focused on isometric holds, such as hollow rocks and Roman chair exercises. Exercise sequencing was varied to promote participant autonomy and well-being. Fatigue and quality were managed by avoiding back-to-back exercises targeting the same muscle group. Trainers ensured proper exercise order and intensity. Training loads adhered to guidelines from the ACSM and NSCA for cerebral palsy. The 12-week program was divided into three 4-week blocks see table 3 from Valadao et al., 2021 for specific information. 1. Block 1: Three sets of eight repetitions at 50% 1RM, with 3-second concentric and eccentric phases, 60 seconds rest between sets, and no rest between repetitions. 2. Block 2: Same volume, but with increased intensity—concentric duration reduced to 1 second, and rest extended to 90 seconds. 3. Block 3: Increased sets (four), reduced repetitions (six), and higher intensity with rapid concentric movements and 2-second eccentric phases. This progression aimed to improve strength, motor control, and power output. Exercise plans, progression details, and session structures are summarized in an accompanying table 3 below. TABLE 3. Training progression from (Valadão et al., 2021). Gait training was conducted using a portable mechanical treadmill with an incline of 6° or 7.3° (Vida XL, Venlo, Netherlands). Participants were instructed to walk at a comfortable pace, avoiding toe walking and aiming for heel strike. A non-motorized treadmill was chosen because its belt moved through hip extension and ankle plantarflexion, promoting proper muscle activation patterns. Verbal feedback was provided during the warm-up to improve gait quality, with participants allowed to stop and rest as needed. Each session included 5-10 minutes of gait training, and participants were given a treadmill to use at home, with the expectation of walking at least 10 minutes daily throughout the intervention. Flexibility training involved four sets of 45 seconds of passive-static stretching at the pain threshold for each muscle group identified as tight in the pre-tests. Hip flexors were stretched in the Modified Thomas Test position, with one leg in full hip flexion and the other in hip extension. The trainer applied hip extension torque at the distal thigh, with the knee either fully flexed (two-joint stretch) or relaxed (one-joint stretch). The seated butterfly stretch targeted the hip adductors, with the participant's back supported against a wall, hips externally rotated, knees flexed, and soles of the feet in contact. The trainer applied pressure on the thighs to induce hip abduction. Hamstrings were stretched in supine position by flexing the participant's hip to about 90° and applying knee extension torque to the posterior shank. No stretching was performed on the triceps surae due to two reasons: 1) the difficulty of stretching all muscles bilaterally within a 90-minute session, and 2) the inclined gait training already provided a passive dynamic stretch to the ankle joint. Therefore, the focus was on the other muscle groups to assess the impact of the gait intervention on flexibility. 4.4 Passive muscle stretch measurement protocol The passive muscle stretch measurement protocol was performed a total of four times with two pretests (Pre1 and Pre2) and two post-tests (Post1 and Post 2), spaced three months apart with two three-month control periods between pre1 and pre2 and post 1 and post 2. It used the following protocol to collect EMG data of the soleus and medial gastrocnemius was evaluated using the TSRT and calculated using the previously described SRT latency corrected method suggested by (Valadão et al., 2022). The protocol was performed a total of four times with two pretests (Pre1 and Pre2) and two post-tests (Post1 and Post 2), spaced three months apart with two three-month control periods between pre1 and pre2 and post 1 and post 2 . The participant was positioned with the knee fully extended, and stretches were applied to the triceps surae by a motor-driven dynamometer (Neuromuscular Research Center, University of Jyväskylä, Finland). The range of motion of passive ankle dorsiflexion was from 20° of plantarflexion to 0° at four angular velocities (55, 110, 210, and 291°/s) The velocities were applied in a pseudorandomized and balanced sequence, with 10 stretches conducted at each velocity and a rest interval of 2.6–2.9 seconds between stretches (Valadão et al., 2021). EMG data were collected and analysed to identify the joint angle at which the TSRT occurred. In this case study the SRT latency corrected method was used to include an individual h-reflex latency correction to identify the joint angle, as suggested in the aims of the EXECP study. This case study used the collected EMG data from the passive muscle stretch protocol from EXECP to calculate the TSRTcorrected. 4.5 EMG data collection and setup The EMG data collection was performed a total of four times with two pretests (Pre1 and Pre2) and two post-tests (Post1 and Post 2), spaced three months apart with two three-month control periods between pre1 and pre2 and post 1 and post 2. For both the passive muscle stretch protocol and the h-reflex protocol. In this study EMG activity was recorded from the Sol, MG muscles using self-adhesive electrodes (Blue Sensor N, Ag/AgCl, 0.28 cm²; Ambu, Ballerup, Denmark), with a ground electrode placed on the tibia. Electrode placement and skin preparation were conducted following SENIAM guidelines previously mentioned by (Hermens et al., 2000). The electrodes were positioned on the muscle belly, aligned with the underlying fibre direction, with a 20-mm interelectrode distance. EMG signals were amplified and high pass filtered (1000X, 10 Hz) using a preamplifier (NL824; Digitimer, Welwyn Garden City, UK) and subsequently bandpass filtered (10 Hz to 1 kHz) by a differential amplifier (NL900D/NL820A; Digitimer Ltd., UK). The signals were acquired on a personal computer at a sampling rate of 5 kHz using a 16-bit A/D converter (CED Power 1401; Cambridge Electronics Design, Cambridge, UK). This case study analysed the EMG data collected from the Sol and MG, to identify the TSRT using Spike2 from Cambridge Electronics Design, Cambridge, UK (Valadão et al., 2021). 4.6 H-reflex recruitment protocol and setup The H-reflex recruitment protocol was performed to identify the H-reflex and outline the recruitment curve. The protocol was performed a total of four times with two pretests (Pre1 and Pre2) and two post-tests (Post1 and Post 2), spaced three months apart with two three-month control periods between pre1 and pre2 and post 1 and post 2 (Valadão et al., 2021). H-reflexes and M-waves were evoked in the soleus and medial gastrocnemius muscles by percutaneous electrical stimulation of the tibial nerve while the participant lay in a prone position. A single rectangular pulse (1 ms) was delivered using a constant-current stimulator (DS7AH; Digitimer, UK). A circular cathode with a pickup area of 0.77 cm² (Unilect4535M, Ag/AgCl; Unomedical, Redditch, UK) was placed over the tibial nerve at the popliteal fossa, and an oval-shaped, 5.1 × 10.2 cm anode (V-trodes; Mettler Electronics, Anaheim, CA) was positioned above the patella. The stimulation site providing the greatest amplitude of evoked responses in SOL and MG was initially identified using a handheld cathode electrode, which was later replaced by a self-adhesive electrode. The full recruitment curve was obtained by starting with a stimulation intensity of 3 mA and increasing in 0.5–2 mA steps (using smaller steps near the maximum Hreflex) at 0.1 Hz until the maximal M-wave was achieved. Peak-to-peak values for the maximum H-reflex and M-wave were then computed (Valadão et al., 2021). 4.7 H-reflex latency calculation protocol The H-reflex latency calculation protocol was as such, H-reflex latency initially calculated using MATLAB (R2023a, The MathWorks Inc, Natick, United States). H-reflex latency calculation was performed using the following steps see the figure 14 below. The first red circle represents the stimulation point, while the m-wave spans the second magenta asterisk to the third magenta asterisk. The H-reflex is identified by the first peak in EMG activity. As shown in the figure 14, the fluorescent green circle marks the start of the H-reflex. To calculate latency, the difference in the position from the stimulation point (red circle) to the fluorescent green point was calculate. FIGURE 14. Individual H-reflex latency calculation example for the case study subject. The red circle is the stimulation point the green circle the beginning of the h-reflex. 4.8 Data analysis preparation Stretch reflex threshold EMG data was for this case study was analysed using Spike2 software from Cambridge Electronics Design, Cambridge, UK. The relevant data channels—medial gastrocnemius (4), soleus (3), and angle (2) were selected. The angle axis was zeroed for each file to ensure values ranged between 20 and 0 degrees. This was done through the channel information screen by entering the appropriate axis offset value. This process was repeated for all data sets to standardize the angle range across the data sets and allow comparison of results. 4.9 Filtering of data in spike The medial gastrocnemius and soleus channels were filtered using the IIR filter function in Spike2 (Cambridge Electronics Design, Cambridge, UK). A custom filter function was created using the band pass fourth order filter function. A Butterworth fourth order filter, with a low of 20Hz and high of 195Hz (20-195Hz) was used as previously suggested by literature and as per other EXECP analysis (Valadão et al., 2021). The filter was applied to both the medial gastrocnemius and soleus channels. New memory channels were created by clicking the apply digital filter and it was saved as RealWave (Real, 4 bytes per point). The channels were renamed as MG for medial gastrocnemius and SOL for soleus to avoid confusion during analysis. The original MG and SOL channels were removed from the display window by right clicking and deselecting channels 3 and 4. The axes for MG and SOL were then optimized to display data on the same scale, ensuring comparability between the two channels. With these preparations complete, data analysis could proceed for the TSRTcorrected, as detailed in the following section. 4.10 The tonic stretch reflex threshold corrected analysis The EMG data from the passive muscle stretching protocol from EXECP was used for this case study to calculate TSRTcorrected. TSRTcorrected was analysed in Spike2 from Cambridge Electronics Design, Cambridge, UK. First the steps in the above section were performed to filter the data. Once this was completed the TSRTcorrected data analysis commenced. Each time point (PRE1,PRE2,POST1 and POST2) was analysed individually and manually, for the 40 stretches. 10 stretches were analysed at each of the four velocities (55,110,200 and 291). Signals were not considered for analysis and were discarded if the signal to noise ratio was too high. This was chosen based on laboratory specific conditions from the NMRC at the university of Jyväskylä. This was done using the right click function, selecting the relevant Hcursor1, set position and entering the values. The TSRTcorrected was analysed by placing two vertical cursors were initially used to determine the stretch velocity. One cursor was placed at the start of the angle slope, and the other at the end. The positions of the cursors were then copied and subtracted from each other to calculate the stretch duration in seconds(s). Four distinct values were identified, corresponding to different velocities: 0.4s (55 º/s), 0.2s (110 º/s), 0.1s (200 º/s), and 0.68s (291 º/s). This process was repeated for each stretch to ensure consistency in the results. After identifying the stretch velocities, the TSRT-corrected EMG analysis was performed. Vertical cursors were used to determine the TSRT onset via EMG burst activity, labelled as SRTNON for non-corrected and SRTCORR for the latency corrected positions. The SRTNON cursor position was copied into the Windows calculator, and the individual H-reflex latency for the muscle was subtracted from this value. The result was then used to adjust the SRTCORR cursor position for precise identification of the TSRTcorrected position. A horizontal cursor was placed at the x, y intercept of the SRTCORR vertical cursor to mark the TSRTcorrected position, representing the joint angle of TSRTcorrected. This horizontal cursor value was then copied into the Excel file for analysis. The process was repeated for both the SOL and MG muscles across all 40 stretches and the four time points (PRE1, PRE2, POST1, POST2). Care was taken to use the relevant H-reflex latency for each time point and muscle, as calculated in MATLAB (version R2023a, The MathWorks Inc, Natick, United States), following the latency correction procedure described earlier. 4.11 Statistical analysis Since this was a case study no complex statistical analysis was performed instead the results were presented in the form of descriptive statistics. This included mean, standard deviation and coefficients of variation for each measurement point these were calculated using Microsoft excel (Microsoft Corporation, 2024). Mean relative change was also calculated to see the relative percentage of change between time points to allow for a slightly more in depth look at the potential changes over time due to the exercise intervention. 5 RESULTS 5.1 Pre-intervention results The Pre 1 results in Table 4 highlight mean latency-corrected stretch reflex threshold characteristics for the soleus (SOL) and medial gastrocnemius (MG) muscles. For the SOL, reflex thresholds increased with velocity, peaking at 15.31° at 200°/s before decreasing slightly to 13.02° at 291°/s. Variability (CV%) also increased with velocity, ranging from 4.9% at 110°/s to a peak of 16.54% at 291°/s, indicating less consistent reflex responses at higher speeds. In contrast, the MG exhibited stable reflex thresholds across velocities, with mean values ranging from 19.36° at 55°/s to 18.91° at 291°/s. The MG also demonstrated remarkably low variability, with CV values progressively decreasing from 3.05% at 55°/s to 0.99% at 291°/s, reflecting highly consistent reflex activation. TABLE 4. Pre1 TSRTcorrected joint angle data mean, standard deviation (SD) and coefficient of variation (CV), for soleus (SOL) and medial gastrocnemius (MG). Muscle Velocity (°/s) Mean SD CV (%) SOL 55 NA NA NA SOL 110 9.59 0.47 4.9 SOL 200 15.31 1.52 9.93 SOL 291 13.02 2.15 16.54 MG 55 19.36 0.59 3.05 MG 110 19.38 0.22 1.15 MG 200 18.90 0.23 1.22 MG 291 18.91 0.19 0.99 The PRE2 data in Table 5 below presents the latency-corrected TSRT for the soleus (SOL) and medial gastrocnemius (MG) muscles across different velocities. In the SOL, the stretch reflex threshold decreased slightly as velocity increased, from 16.69° at 110°/s to 15.54° at 291°/s. Variability (CV%) was highest at 200°/s (10.13%) and slightly lower at 291°/s (9.84%), indicating greater inconsistency at higher velocities. For the MG, the latency-corrected TSRT remained stable across velocities, with mean values ranging from 17.56° at 55°/s to 18.79° at 110°/s. Variability (CV%) was highest at the lowest velocity, 55°/s (18.82%), and significantly lower at higher velocities, reaching its lowest at 110°/s (2.94%). The MG showed greater consistency across velocities, with variability decreasing as velocity increased, reflecting more reliable stretch reflex activation at higher speeds. The findings from this case study do help to reveal distinct functional roles and neuromechanical properties of the soleus and medial gastrocnemius muscles, evident in their TSRTcorrected joint angle profiles. The soleus had lower mean joint angles, indicating earlier reflex activation and greater sensitivity to stretch, aligning with its primary function in postural stability and controlled movement. This sensitivity is supported by its composition of slowtwitch (type I) fibres, which enable rapid stabilization during standing and walking to maintain balance and counteract perturbations (Sinkjaer et al., 1996). Conversely, the medial gastrocnemius showed higher mean joint angles at the TSRT, indicating a delayed reflex onset. This delay facilitates high-speed, forceful movements like running and jumping by allowing greater joint motion for optimal force generation before reflex activation. Its higher proportion of fast-twitch (type II) fibres reinforces its role in dynamic, explosive tasks (Armand et al., 2016). It is well established that resistance training, when appropriately structured, can promote muscle hypertrophy and enhance force and power in children and adolescents with CP. These interventions may also play a critical clinical role in increasing fascicle length, which is important for reducing contractures and improving muscle torque–angle and torque–velocity properties (Moreau & Lieber, 2022). However, the precise mechanisms driving changes in muscle size and fascicle length are not fully understood. Future research with adequately powered studies and large sample sizes is needed to address these uncertainties, enabling clinicians to design targeted rehabilitation protocols to achieve specific muscle functional outcomes (Moreau & Lieber, 2022). TSRTcorrected variability also differed between the muscles. The soleus exhibited greater variability, particularly at moderate velocities (e.g., 200°/s, CV = 10.12%), reflecting its adaptability to diverse postural and dynamic demands. This variability balances its stabilizing role with responsiveness to changing mechanical conditions. In contrast, the medial gastrocnemius displayed low variability across all velocities, with the most consistent responses observed at 291°/s (CV = 1.07%). This consistency highlights its reliability during high-speed movements, where precise reflex activation is critical. Velocity influenced reflex activation patterns differently in each muscle. For the soleus, reflex threshold joint angles peaked at 200°/s and decreased at 291°/s. This suggests an adaptive mechanism at moderate velocities that permits greater joint motion, while earlier activation at higher velocities helps prevent overstretching and maintains stability (Sinkjaer et al., 1996). The medial gastrocnemius exhibited consistent reflex thresholds across velocities (17.53°–18.01°), demonstrating its ability to support rapid movements without significant changes in reflex sensitivity. This stability aligns with the muscle’s greater stiffness and faster contraction capabilities (Armand et al., 2016). To summarise the relative mean change data from table 6 suggests that there may have been a positive effect on TSRTcorrected mean joint angle due to the exercise intervention between the PRE2 and POST 1 measurements. Between the POST 1 and POST 2 time point the values returned towards baseline for the 110°/s and 291°/s SRT’s suggesting a normalisation or reduction in the beneficial effects of the intervention once it ceased in this case study subject. Interestingly, the 200°/s SRT mean showed a continued decrease of -6% suggesting further reduction in the SRT at the POST2 measurement. These values must be interpreted cautiously as they have not undergone statistical analysis. However, they provide some promise regarding the effectiveness of the EXECP intervention on reducing SRT which may be beneficial for stretch hyper-reflexia. The post intervention period may have also seen a reduction in the benefits, but some residual effects may be leftover. To confirm these a more rigorous statistical analysis should be performed on the study cohort. Overall, the soleus is characterized by heightened sensitivity and variability, suited to lowintensity, controlled tasks. In contrast, the medial gastrocnemius prioritizes stability and delayed reflex activation, optimized for high-speed, dynamic activities. The soleus activates early to stabilize and prevent falls during slow movements (Sinkjaer et al., 1996), while the medial gastrocnemius’s delayed reflex reduces interference, enabling powerful performance in activities such as sprinting and jumping (Kalkman et al., 2018). 6.1 Potential effects of the exercise intervention Several potential mechanisms could explain the changes observed in muscle response following the exercise intervention. These mechanisms would likely relate to neural adaptations, muscle fibre remodelling, changes in muscle stiffness, and modifications in the stretch reflex pathway. The soleus exhibited heightened sensitivity to stretch, which could be due to enhanced proprioceptive input and better coordination between the muscle spindles and motor neurons. Exercise, particularly those focused on balance and postural control (resistance or functional training), can improve neural efficiency, enhancing the ability of the muscle to respond quickly to perturbations. Exercise can lead to CNS adaptations that improve motor control and response times. For example, increased neural drive or improved synchronization of motor units could contribute to the earlier reflex activation seen in the soleus, helping to stabilize the body during dynamic movements like walking or standing (Enoka, 2015). The soleus, which is predominantly composed of slow-twitch fibres, is important in maintaining posture. Exercise interventions, especially those involving endurance or low-intensity tasks, may increase the proportion or the endurance of type I fibres, leading to improved performance in stabilization tasks. These fibres are more sensitive to reflex activation, supporting the observed results. In the medial gastrocnemius, which is primarily composed of type II fibres, high-intensity exercise, particularly explosive or sprint training, could lead to an increase in the number or size of these fibres (McArdle et al., 2010). This would result in enhanced force production, supporting the muscle’s delayed reflex activation, as greater joint motion would be needed to generate maximal force before the reflex response occurs (A. J. Blazevich, 2006). The most recent paper from Valadao et al., 2024, showed that the EXECP intervention significantly improved thigh muscle strength. Knee extensors (KE) showed a 19% increase in peak torque, 31% in the rate of force development, and 24% in concentric angular impulse. Knee flexors (KF) exhibited even greater gains, with a 31% rise in peak torque and a 92% increase in concentric angular impulse. These gains align with previous studies reporting 12%– 27% improvements in KE torque and 25% increases in concentric torque. KF improvements likely reflect enhanced motor control as well as strength. However, shank muscle strength (PF and DF) did not improve significantly, possibly due to suboptimal training loads and testing protocols that limited adaptation. Trunk muscle strength improvements followed expected patterns. For extensors, load increases reached 900%, as 12 participants achieved maximum exercise duration (60 s) early in the intervention. Flexor strength gains involved a 33% increase in hold duration and weight adjustments, which were more challenging due to long resistance arms. Curve width, indicating steady torque output, did not improve for any muscle group, likely due to the intervention's initial training load structure. Overall, psychological factors, such as willingness to exert maximum effort, may have influenced strength gains, as suggested by the 10-fold increase in trunk extension weight (Valadão et al., 2024). Muscle and or tendon stiffness can influence the stretch reflex threshold, with more stiffness contributing to a higher threshold for reflex activation (Lieber & Fridén, 2019). Exercise interventions that focus on strength training or plyometrics can increase the stiffness of the muscle-tendon unit, which may contribute to the medial gastrocnemius’s consistent stretch reflex threshold at higher velocities (A. Blazevich & Blazevich, 2017). A stiffer muscle can better resist rapid stretching and will need greater joint motion to activate the reflex. Tendon stiffness, which can also be modified by exercise, may affect how quickly muscles are stretched and how efficiently they generate force. Increased tendon stiffness may reduce the amount of stretch needed to activate the reflex in the medial gastrocnemius, while also providing more efficient force transmission during dynamic movements like running and jumping (Enoka, 2015). The most recent paper from Valadao et al., 2024, showed that the EXECP intervention improved joint flexibility by a mean of 5° in passive knee and hip extensions, a statistically significant but functionally minimal change achieved with just 6 minutes of weekly training. It remains unclear whether the gains stemmed from sensorimotor adaptations (increased stretch tolerance) or structural changes in the muscle-tendon unit. Evidence suggests that flexibility training alone is unlikely to induce mechanical changes in individuals with CP due to limited muscle growth capacity and hindered sarcomere genesis. Given the lack of strong evidence supporting stretching for CP, therapeutic focus should shift toward proven interventions like strength and gait training. Importantly, the intervention showed no negative impact on joint flexibility, addressing a common concern among health practitioners (Valadão et al., 2024). Exercise interventions, particularly those that emphasize coordination or proprioception could enhance the integration of sensory inputs (from muscle spindles, Golgi tendon organs) and motor outputs. This could lead to improved timing and modulation of the stretch reflex, as seen in the soleus, which showed earlier reflex activation. The exercise could lead to changes in the stretch reflex threshold, influenced by adaptations in spinal cord processing. With regular, targeted training, reflex thresholds may shift to allow for more optimal muscle activation patterns, which could explain the reduced variability in the medial gastrocnemius and its ability to maintain consistent reflex thresholds (Lundy-Ekman & Weyer, 2022). Exercise could lead to more efficient recruitment of motor units in both the soleus and medial gastrocnemius. This may manifest as a faster response time (for the soleus) or more controlled activation patterns (for the medial gastrocnemius), allowing for the observed changes in reflex sensitivity and threshold (Enoka, 2015). The variability in reflex thresholds and timing could also be influenced by the velocity of movement, as observed in the soleus. Exercises that challenge the muscles at different movement speeds, such as resistance training with varying tempos or agility drills, could enhance the muscle’s ability to modulate its reflex response at different velocities, contributing to the observed peak in reflex thresholds at moderate speeds and decrease at higher speeds (Valadão et al., 2022). Training interventions focusing on balance, such as standing on unstable surfaces, could enhance the sensitivity of the soleus for maintaining postural stability. Increased proprioceptive input, along with more efficient reflex activation, could explain why the soleus showed an earlier reflex onset and greater variability at moderate speeds likely a result of the muscle’s adaptive role in dynamic postural control (Sinkjaer et al., 1996). The medial gastrocnemius’s delayed reflex activation could be linked to an exercise-induced adaptation that optimizes muscle control in explosive movements. High-velocity training that involves jumping, sprinting, or other fast, powerful movements could improve the muscle's ability to generate force before reflex activation kicks in, leading to a more efficient performance during dynamic actions (McArdle et al., 2010). Furthermore, the most recent paper published by Valadao et al., 2024 found the following results. The EXECP study observed a significant improvement in motor function, with the group mean 6MWT distance increasing by 33 m (Post1) and 44 m (Post2) compared to pretests. These changes exceed the typical test-retest variability (±43 m), indicating clinical significance. Similar results were seen in Gillet et al.'s study with younger participants (mean age 20 years) combining strength and anaerobic training, while Kirk et al.’s strength-only program for older participants (mean age 36 years) did not improve 6MWT despite gains in gait kinematics and muscle strength. Differences in training protocols, particularly the inclusion of motor coordination exercises, may explain these disparities. Regarding GMFM scores, nine participants achieved a clinically significant improvement of at least two points, with functional gains such as climbing stairs with alternating feet. However, smaller improvements, like balancing on one foot for a few seconds longer, may be less functionally impactful. This underscores the importance of evaluating both quantitative and qualitative outcomes to determine the relevance of such changes (Valadão et al., 2024). In summary, the observed changes in stretch reflex thresholds and muscle activation are likely due to a combination of neural adaptations, changes in muscle fibre composition and stiffness, improved sensory-motor integration, and optimized muscle recruitment strategies. Exercise interventions tailored to specific functional needs such as postural stability for the soleus or explosive movement for the medial gastrocnemius could drive these mechanisms, leading to the enhanced performance seen in the results. 6.2 Strengths and weaknesses of the study The strengths and weaknesses of this study are varied. A key strength of this study is that it is one of the first of its kind to assess the latency corrected TSRT at several measurement points. That is its longitudinal nature and allows for more insight into the effectiveness of the exercise intervention immediately pre and post intervention and at a three-month follow-up. A clear weakness is the fact that this was a case study and as such no strong conclusion can be drawn from the results, they may only help give us an idea of potential outcomes related to the exercise intervention and its effect on the TSRT. Another area which could be improved is the analysis of the latency corrected TSRT. It could be valuable to use multiple evaluators to assess the TSRT in this way to see what kind of repeatability there is in the methodology. How does the experience of the evaluator influence their identification of the TSRT. Some guidelines related to false positives or false negative results could also help to improve reliability when assessing the TSRT. Some other strengths of the study are the fact that the latency corrected TSRT was assessed at several time points and velocities. Which help it to identify whether there is a velocity dependent relationship or not. Furthermore, a more comprehensive statistical analysis procedure could help us highlight any significant results. Thus, highlighting whether the intervention was effective in reducing TSRT in spastic cerebral palsy. 7 CONCLUSION In summary, the medial gastrocnemius shows greater consistency in both its stretch reflex threshold and variability across a range of velocities. This suggests that the medial gastrocnemius may be better suited for tasks requiring stable performance at varying speeds. In contrast, the soleus exhibits greater variability, particularly at moderate velocities, but tends to stabilize at higher speeds, likely reflecting its specialized function in maintaining postural stability during slow or controlled movements, which require more adaptability to changes in movement dynamics. These findings highlight the distinct functional roles of these muscles in dynamic activities and could inform approaches to rehabilitation or training that target specific velocity-dependent muscle behaviours. The soleus exhibited greater sensitivity and variability in stretch reflex activation, aligning with its postural and stabilizing role. In contrast, the medial gastrocnemius demonstrated stable and reliable reflex thresholds, supporting its function in dynamic, high-force movements. The exercise intervention appeared to improve reflex consistency in the soleus while maintaining stability in the medial gastrocnemius. However, as no statistical analysis was performed, we must be careful interpreting them as such, and further in-depth analysis is necessary. These findings emphasize the distinct neuromechanical roles of these muscles and the potential for targeted interventions to enhance motor control in individuals with cerebral palsy who exhibit spasticity. 8 REFERENCES Barber, L., Barrett, R., & Lichtwark, G. (2012). Medial gastrocnemius muscle fascicle active torque-length and Achilles tendon properties in young adults with spastic cerebral palsy. Journal of Biomechanics, 45(15), 2526–2530. https://doi.org/10.1016/j.jbiomech.2012.07.018 Bar-On, L., Kalkman, B. M., Cenni, F., Schless, S.-H., Molenaers, G., Maganaris, C. N., Bass, A., Holmes, G., Barton, G. J., O’Brien, T. D., & Desloovere, K. (2018). The Relationship Between Medial Gastrocnemius Lengthening Properties and Stretch Reflexes in Cerebral Palsy. Frontiers in Pediatrics, 6, 259. https://doi.org/10.3389/fped.2018.00259 Blazevich, A., & Blazevich, A. J. (2017). Sports biomechanics: The basics: Optimising human performance. Bloomsbury Publishing. Blazevich, A. J. (2006). Effects of physical training and detraining, immobilisation, growth and aging on human fascicle geometry. Sports Medicine, 36, 1003–1017. Castle, M. E., Reyman, T. A., & Schneider, M. (1979). Pathology of spastic muscle in cerebral palsy. Clinical Orthopaedics and Related Research, 142, 223–233. https://doi.org/10.1097/00003086-197907000-00036 Enoka, R. M. (2015). Neuromechanics of human movement 5th edition. Human Kinetics. Graham, H. K., Rosenbaum, P., Paneth, N., Dan, B., Lin, J.-P., Damiano, D. L., Becher, J. G., Gaebler-Spira, D., Colver, A., Reddihough, D. S., Crompton, K. E., & Lieber, R. L. (2016a). Cerebral palsy. Nature Reviews Disease Primers, 2(1), 15082. https://doi.org/10.1038/nrdp.2015.82 Graham, H. K., Rosenbaum, P., Paneth, N., Dan, B., Lin, J.-P., Damiano, D. L., Becher, J. G., Gaebler-Spira, D., Colver, A., Reddihough, D. S., Crompton, K. E., & Lieber, R. L. (2016b). Cerebral palsy. Nature Reviews Disease Primers, 2(1), 15082. https://doi.org/10.1038/nrdp.2015.82 Hermens, H. J., Freriks, B., Disselhorst-Klug, C., & Rau, G. (2000). Development of recommendations for SEMG sensors and sensor placement procedures. Journal of Electromyography and Kinesiology, 10(5), 361–374. Howard, J. J., & Herzog, W. (2021). Skeletal Muscle in Cerebral Palsy: From Belly to Myofibril. Frontiers in Neurology, 12, 620852. https://doi.org/10.3389/fneur.2021.620852 Kalkman, B. M., Bar-On, L., Cenni, F., Maganaris, C. N., Bass, A., Holmes, G., Desloovere, K., Barton, G. J., & O’Brien, T. D. (2018). Muscle and tendon lengthening behaviour of the medial gastrocnemius during ankle joint rotation in children with cerebral palsy. Experimental Physiology, 103(10), 1367–1376. https://doi.org/10.1113/EP087053 Kalsi, G., Fry, N. R., & Shortland, A. P. (2016). Gastrocnemius muscle–tendon interaction during walking in typically-developing adults and children, and in children with spastic cerebral palsy. Journal of Biomechanics, 49(14), 3194–3199. https://doi.org/10.1016/j.jbiomech.2016.07.038 Levin, M. F., Piscitelli, D., & Khayat, J. (2024). Tonic stretch reflex threshold as a measure of disordered motor control and spasticity – A critical review. Clinical Neurophysiology, 165, 138–150. https://doi.org/10.1016/j.clinph.2024.06.019 Lieber, R. L., & Fridén, J. (2019). Muscle contracture and passive mechanics in cerebral palsy. Journal of Applied Physiology, 126(5), 1492–1501. https://doi.org/10.1152/japplphysiol.00278.2018 Lundy-Ekman, L., & Weyer, A. (2022). Neuroscience: Fundamentals for Rehabilitation. Elsevier. https://books.google.com.au/books?id=82GkzgEACAAJ Matthiasdottir, S., Hahn, M., Yaraskavitch, M., & Herzog, W. (2014). Muscle and fascicle excursion in children with cerebral palsy. Clinical Biomechanics, 29(4), 458–462. McArdle, W. D., Katch, F. I., & Katch, V. L. (2010). Exercise physiology: Nutrition, energy, and human performance. Lippincott Williams & Wilkins. Meyns, P., Van Gestel, L., Leunissen, I., De Cock, P., Sunaert, S., Feys, H., Duysens, J., Desloovere, K., & Ortibus, E. (2016). Macrostructural and Microstructural Brain Lesions Relate to Gait Pathology in Children With Cerebral Palsy. Neurorehabilitation and Neural Repair, 30(9), 817–833. https://doi.org/10.1177/1545968315624782 Moreau, N. G., & Lieber, R. L. (2022). Effects of voluntary exercise on muscle structure and function in cerebral palsy. Developmental Medicine & Child Neurology, 64(6), 700– 708. https://doi.org/10.1111/dmcn.15173 Mukherjee, A., & Chakravarty, A. (2010). Spasticity mechanisms–for the clinician. Frontiers in Neurology, 1, 149. Nielsen, J. B., Petersen, N. T., Crone, C., & Sinkjaer, T. (2005). Stretch Reflex Regulation in Healthy Subjects and Patients with Spasticity. Neuromodulation: Technology at the Neural Interface, 8(1), 49–57. https://doi.org/10.1111/j.1094-7159.2005.05220.x Pagnozzi, A. M., Dowson, N., Doecke, J., Fiori, S., Bradley, A. P., Boyd, R. N., & Rose, S. (2017). Identifying relevant biomarkers of brain injury from structural MRI: Validation using automated approaches in children with unilateral cerebral palsy. PLoS One, 12(8), e0181605. Peeters, N., Hanssen, B., Bar-On, L., De Groote, F., De Beukelaer, N., Coremans, M., Van den Broeck, C., Dan, B., Van Campenhout, A., & Desloovere, K. (2023). Associations between muscle morphology and spasticity in children with spastic cerebral palsy.