The syndrome of deforming spastic paresis

pathophysiology, assessment and treatment

Central nervous system (CNS) lesions affecting central motor pathways involved in motor command execution lead to the syndrome of deforming spastic paresis, i.e., stretch-sensitive paresis associated with soft tissue shortening and stretch-sensitive muscle overactivity (Gracies, 2005a,b). Such lesions may be caused by disorders as common as infant paresis (a term we feel is more appropriate than cerebral palsy) but also adult disorders such as stroke, traumatic brain injury, multiple sclerosis, spinal cord injury, anoxic brain injury, primary lateral sclerosis and hereditary spastic paraparesis.

We proposed a simplification of the Lance et al. (1980) definition of spasticity as an ‘increase in the velocity-dependent reflexes to phasic stretch’ (Gracies, 2005b) This increase, measured at rest, is manifested by both a reduced threshold and an increased gain of the responses to stretch (Gracies, 2005b). As such, spasticity is probably not the most disabling type of muscle overactivity. The whole syndrome of deforming spastic paresis, however, causes limb deformities and motor limitations hindering social life, mobility and activities of daily living. This chapter reviews the triple phenomenology of deforming spastic paresis (stretch-sensitive paresis, soft tissue contracture and muscle overactivity) including the definition of the different types of muscle overactivity, methods of clinical evaluation, and therapeutic methods currently established through controlled protocols, particularly as it pertains to children with infant paresis.

Phenomenology and taxonomy in deforming spastic paresis

Three fundamental phenomena occur after a lesion to the central pathways involved with motor command execution: stretch-sensitive paresis, soft tissue contracture and spastic muscle overactivity.

Stretch-sensitive paresis

Paresis, i.e., the quantitative lack of voluntary command accessing agonist muscles when attempting to generate force or movement, is the first, immediate consequence of a CNS lesion involving the corticospinal pathway (Gracies, 2005a). This lack of voluntary activation involves insufficient synchronization, insufficient discharge frequency and higher firing threshold of the recruited motor units. The term stretch-sensitive paresis was coined to refer to further reduction of the ability to recruit motor units in an agonist effort when a contractured, spastic antagonist is stretched (Gracies, 2005b). This phenomenon occurs particularly for the less contractured of two muscles around a joint.

Soft tissue contracture

As a consequence of paresis, some muscles and their surrounding soft tissues are left immobilized in a shortened position, relating to the influence of gravity when patients are left in a specific body posture most of the day (often reclined, or supine). Reciprocally, their antagonists are also submitted to immobilization due to paresis but in a normal or lengthened position. This is the first source of the asymmetry that will characterize muscle changes around joints, with some muscles (often lower limb extensors and upper limb flexors, internal rotators and pronators) undergoing contracture to a much greater extent than their antagonists. In a muscle left immobilized in a short position, muscle plasticity, characterized by gene changes and transcriptional events leading to protein synthesis modifications within muscle fibres (Baptista et al., 2010; Giger et al., 2009), initiates as soon as a few hours after immobilization onset. These modified transcriptional profiles have also been observed and analysed in children with infant paresis (Smith et al., 2009). These muscle changes are understood under the term contracture, which involves at least: (a) physical shortening, which adapts soft tissue (muscles, tendons, ligaments, joint capsules, skin, vessels and nerves) to its newly imposed length; (b) reduced extensibility; and (c) modifications of muscle contractile properties, e.g., slow-to-fast changes of originally slow muscles (Baptista et al., 2010; Giger et al., 2009; Gracies, 2005a; Smith et al., 2009). These early transformations only intensify in the days and weeks that follow the onset of immobilization, if no or insufficient preventative treatment is implemented (Gracies, 2005b). They initiate the body deformities characteristic of deforming spastic paresis. When the initial neural paresis-causing lesion has occurred in the perinatal period, muscle contracture will later affect bone growth along with normal muscle lengthening (Hof, 2001). With today’s therapeutic management, it is likely that such contracture is not adequately treated in paretic children as a large retrospective study in plantar flexor muscles recently showed that the range of ankle dorsiflexion decreases by 19 degrees during the first 18 years of life in children with infant paresis (Hägglund and Wagner, 2011). These disfigurements will become increasingly challenging, socially and psychologically, if appropriate preventative or curative treatment is not implemented.

Spastic muscle overactivity

After damage to corticospinal pathways, lesion- and behaviour-induced adaptive changes take place within higher centres and the spinal cord (Gracies, 2005b). Intraspinal reorganization occurs as growth factors, adhesion and guidance molecules and other components fostering synapse neogenesis are produced by denervated ventral horns at spinal segments (Giger et al., 2010; Maier et al., 2008; Raineteau et al., 2002). Mature or maturing descending motor tracts (brainstem descending and contralesional corticospinal pathways) thus undergo intraspinal reconnections, through local cues guiding newly formed sprouts in the denervated spinal cord (Maier et al., 2008; Raineteau et al., 2002). Brainstem descending pathways (rubro-, tecto-, reticulo-, vestibulospinal) and contralesional corticospinal pathways are increasingly recruited at higher centres—potentially via frontal or transcallosal disinhibition after brain lesions—to take over some of the motor command execution (Ghosh et al., 2009; Giger et al., 2010; Maier et al., 2008; Raineteau et al., 2002; Riddle and Baker, 2010 ). Most of these brainstem descending pathways tend to be permanently active at rest, thus creating the conditions for the emergence of permanent, dystonic muscle activity through their newly formed motoneuronal connections, as such ongoing descending activity impacts on a sensitized, hyperexcitable motor neuron (‘denervation hypersensitivity’, see below).

At each spinal level, local sprouting from neighbouring interneurons is also promoted, fostering the formation of new abnormal synapses between these interneurons and the somatic membrane of the deprived motor neurons. These new segmental or propriospinal synapses form a substratum for the emergence of motor neuron hyperexcitability and new abnormal or exaggerated reflex pathways (Gracies, 2005b). The superimposition of augmented descending and reflex inputs on hyperexcitable motor neurons leads to overall muscle overactivity, which adopts various forms.

Spasticity

Spasticity has been the most commonly recognized manifestation among these gradually occurring reflex changes. Using a simplified definition as an increase in the velocity-dependent stretch reflexes (Gracies, 2005b), it is possible to observe spasticity at rest by excessive responses to muscle stretch or tendon taps. When compared to normal subjects, stretch-induced contraction at rest occurs at lower threshold and with increased amplitude in patients with deforming spastic paresis.

Thus, the optimal condition for detecting and evaluating spasticity is the use of phasic stretch (i.e., the movement of stretch) at rest. Spasticity does not constitute a highly disabling form of muscle overactivity, except in attempts at fast or ballistic active movements or when stretch-induced clonus is triggered and interferes with posture or movement in activities such as nursing or dressing.

Spastic dystonia

From his animal studies using motor or premotor cortex ablations, Denny-Brown has coined the term spastic dystonia to describe the tonic, chronic muscle activity present at rest in the context of spasticity, a symptom confirmed by Laplane in humans affected by similar lesions (Denny-Brown, 1966; Gracies, 2005b; Laplane et al., 1977). Spastic dystonia represents spontaneous muscle overactivity at rest, without a specific triggering factor. A simple observation of patients with deforming spastic paresis at rest may often point to this type of muscle overactivity, as it enhances deformity of joint and body postures. Spastic dystonia thus represents an additional major cause of disfigurement and social disability, as it exacerbates the cosmetic consequences of soft tissue contracture.

Thus, the optimal condition for detecting and evaluating spastic dystonia is natural observation at rest, without phasic or tonic stretch. Simple experiments have demonstrated actual reduction of the degree of spastic dystonia with sustained stretch (Gracies and Simpson, 2004). The choice of the term spastic dystonia thus finds double justification: this form of dystonia is present in the context of spasticity and is itself stretch-sensitive.

Spastic co-contraction

Spastic co-contraction is defined as an unwanted, excessive level of antagonistic muscle activity during voluntary agonist command, which is aggravated by stretch of the co-contracting muscle (Gracies, 2005b; Gracies et al., 1997). Spastic co-contraction is a descending phenomenon, most likely due to misdirection of the supraspinal drive during voluntary command (Gracies, 2005b; Gracies et al., 1997; Kukke and Sanger, 2011). It may be facilitated by increased recurrent inhibition, causing loss of reciprocal inhibition to antagonists during voluntary agonist command (Crone et al., 2003; Gracies, 2005b; Gracies et al., 2009; Vinti et al., 2012). Spastic co-contraction is aggravated as effort increases in intensity and duration (Gracies et al., 1997; Vinti et al., 2012).

Thus, the optimal condition for detecting and evaluating spastic co-contraction is voluntary agonist command, regardless of any stretch imposed on muscles. Spastic co-contraction likely represents the most disabling form of muscle overactivity in deforming spastic paresis, also in children with infant paresis, as it impedes force or movement generation, diminishes range of active motion and resists alternating movement (Damiano et al., 2000; Kukke and Sanger, 2011; Vinti et al., 2012).

Other types of muscle overactivity

Other types of muscle overactivity comprise forms that may not be prominently stretch-sensitive. These forms of overactivity include excessive extrasegmental co-contraction, termed ‘synkinesis’, ‘overflow’, ‘associated reactions’, or even ‘athetosis’ or ‘chorea’ depending among others on movement briskness or speed and on the author’s discipline (Chui et al., 2010; Kukke and Sanger, 2011). Excessive extrasegmental co-contraction is an unwanted, abnormally high recruitment of muscles that are distant (at a different segmental level) from the agonist involved in the voluntary command (Gracies, 2005b). Extrasegmental co-contraction may become particularly prominent in children with infant paresis and constitute a separate source of ‘dynamic deformity’ and social and functional disability, which has been correlated with spasticity in one study (Chiu et al., 2010).

Excessive cutaneous or nociceptive responses represent another important form of muscle overactivity, which is often prominent and disabling in some forms of infant paresis or conditions such as spinal cord injury or multiple sclerosis. Finally, inappropriate motor recruitment during autonomic or reflex activities, such as breathing, coughing and yawning, is another characteristic feature in most forms of deforming spastic paresis (Gracies, 2005b).

Clinical evaluation of deforming spastic paresis

The clinical assessment of peripheral paresis involves an ordinal evaluation of each muscle group as an agonist, i.e. for its capacity to generate movement or agonist power, using for example manual measurement with the Medical Research Council scale (John, 1984). This scale provides a number between 0 and 5 meant to represent the amount of command accessing the agonist. Such approach is not valid in deforming spastic paresis, as agonist weakness is impossible to assess manually because of the presence of co-contraction in the antagonist muscle (Crone et al., 2003; Gracies et al., 1997, 2009; Kukke and Sanger, 2011; Vinti et al., 2012).

In direct contrast, in deforming spastic paresis we recommend to assess each muscle group as an antagonist, i.e. for its potential to oppose movement (Gracies et al., 2010a). This strategy derives from Tardieu’s concept that motor impairment in deforming spastic paresis owes more to resistance from hypoextensible soft tissue and excessively activated antagonistic muscles than to lessened agonist command (Tardieu, 1966). We recommend to run a five-step assessment in which the first three steps (slow passive, fast passive and active movement against the muscle tested) measure the potential of muscles to oppose passive and active movements. Each step is quantified by estimating the joint angles at which resistance from the tested antagonist arrests each type of movement assessed (Gracies et al., 2010a). In this strategy, the zero is always defined as the angle of minimal stretch of the tested antagonist (Tardieu scale) (Gracies, 2001a; Gracies et al., 2010a,b; Patrick and Ada, 2006; Tardieu, 1966). The fourth step tests the ability to repeat active performance against the tested muscle. The final step evaluates limb function (Gracies et al., 2010a).

One may rate either objective patient performance as achieved before the rater, or the subjective perception of patient performance, by rater or patient, based on interviews or questionnaires. To objectively rate patient performance, authors have often developed scores of motor impairment, which do not involve real-life tasks (Desrosiers, 1993; Fugl-Meyer et al., 1975; Heller et al., 1987; Lindmark and Hamrin, 1988; Lyle, 1981; Mathiowetz et al., 1985; Rapin et al., 1966). In children with infant paresis, analogous impairment tests include the Bruininks–Oseretsky Test of Motor Proficiency (Bruininks, 1978; Doll, 1946) and the Quality of Upper Extremity Skills Test (QUEST) (DeMatteo et al., 1992). Other scales, however, aim to directly assess real-life activities. In adults, these may be the Frenchay Arm Test, the Rivermead Motor Assessment, the Wolf Motor Function Test, the Jebsen–Taylor Hand Function Test or the Modified Frenchay Scale (Blanton and Wolf, 1999; Collen et al., 1990; Gracies et al., 2002; Jebsen et al., 1969; Lincoln and Leadbitter, 1979; Wade et al., 1983). In children with infant paresis, the Jebsen–Taylor Hand Function Test is also used and specific video-recorded tests have been developed such as the Melbourne Assessment (Johnson et al., 1994) and the Assisting Hand Assessment for unilateral deficiencies (Krumlinde-Sundholm and Eliasson, 2003).

Subjective patient perception on specific upper limb function in adult hemiparesis may be rated in four domains (limb positioning, hygiene, dressing, pain) in the Disability Assessment Scale (DAS), which has shown reliability and usefulness in adult therapeutic studies (Brashear et al., 2002). In children with infant paresis, the Canadian Occupational Performance Measure (COPM) is designed to assess outcomes in the areas of self-care, productivity and leisure, using a semi-structured interview, yielding two scores, for performance and satisfaction with performance (Law et al., 1990). In the Manual Ability Classification System, parents, teachers or the child itself are asked questions on the child’s ability to handle objects in important daily activities: for example during play and leisure, eating and dressing (Eliasson et al., 2006).

Goal Attainment Scaling may be singled out as an attempt at assessing the result of a therapeutic intervention with respect to stated specific goals of the treatment, previously agreed upon between rater and patient, or rater and the patient’s parents (Clark and Caudrey, 1983; Sakzewski et al., 2007). This measure may have good sensitivity to change in children with cerebral palsy (Sakzewski et al., 2007). However, it is often difficult to predict the most important effects that a therapeutic intervention will produce for a given patient; in addition, ‘success’ or ‘failure’ may itself be subjective as it may only depend on how high aims have been set for a given therapeutic intervention. Finally, such assessment is not able to characterize the functional level of a patient but simply whether an intervention has succeeded or not in modifying this level. Therefore, while Goal Attainment Scaling may be interesting as a complement to the functional assessments described above, it probably cannot replace these more systematic assessments.

Lower limb

A major function of the lower limb is ambulation. To objectively rate lower limb function, walking tests are useful (10 m walking test, 2-minute or 6-minute endurance tests), as walking speed has good ecological validity and correlates with most kinematics gait parameters (speed, step length) in hemiparesis (Moseley et al., 2004). In children with infant paresis, stride length, gait velocity and range of motion are related to the severity of the overall physical handicap. During a walking test, step length and cadence may be measured as well as the physiological cost index, which is the speed divided by the difference between the heart rates measured before and after the effort (Butler et al., 1984; Rose et al., 1985). Instrumental kinematic analysis may be proposed at the laboratory as a complement to these assessments, especially when surgery or another specific procedure is considered (Hutin et al., 2010; Skrotzky, 1983). Finally, a number of questionnaire-based subjective functional scales are available (e.g., the Functional Ambulation Classification or the SIP68 mobility subscale) to evaluate ambulation in daily life (Dobkin et al., 2010; Post et al., 1996). The Gross Motor Function Classification System (GMFCS) is a similar classification used for children with infant paresis, defining five levels of self-initiated ambulation capacities from level I (walks without limitations) to level V (transported in a manual wheelchair system) (Palisano et al., 2008). However, while appropriate to characterize the way patients rate their ambulation capacities, such scales may lack sensitivity to therapeutic interventions over short periods (Dobkin et al., 2010; Post et al., 1996).

Management of deforming spastic paresis in children

Treatment of the deformities due to soft tissue shortening: high-load and long-duration stretch

It is important to reiterate here that this treatment, while conceptually simple, is not adequately implemented in most centres today, particularly in children with infant paresis (cerebral palsy) who by the age of 7 years are already characterized by much less extensible muscles than age-matched typically developing children and in whom, as an example, the range of passive ankle dorsiflexion decreases by 19 degrees during the first 18 years of life (Alhusaini et al., 2010; Hägglund and Wagner, 2011). In fact, our personal observation when examining children with cerebral palsy, as they have grown up to be adults, is that their motor impairment owes more to resistance from soft tissue (mechanical) than to central command disorders (neurological), while the opposite is often true for patients with lesions acquired at an adult age.

In deforming spastic paresis, the intertwining between muscle contracture and stretch-sensitive muscle overactivity (Pollock and Davis, 1930; Ranson and Dixon, 1928; Tardieu et al., 1979) contributes to aggravating and consolidating asymmetrical shortening predominating on the more overactive agonist, with the emergence of increasingly fixed deformities (Gracies, 2005b). Here we are reviewing the large body of evidence for the beneficial effect of chronic stretch in deforming spastic paresis, to oppose soft tissue shortening and reduce stretch-sensitive muscle overactivity (Gracies, 2001b).

In animals, chronic stretch has been shown to promote muscle growth by increasing muscle mass, more so than muscle exercise, through genetic, ultrastructural, biochemical (protein content increase), histological changes (weight, length, number of sarcomeres in series, cross-sectional area of type I fibres) and angiogenesis (Carson and Booth, 1998; Cox et al., 2000; Egginton et al., 2001; Goldspink, 1999; Kelley, 1996). Short durations of daily intermittent stretch reversibly increase muscle mass even in muscles otherwise immobilized in short position (Bates, 1993; Carson et al., 1995; Sparrow, 1982; Williams, 1990). Brief muscle stretch also depresses subsequent stretch reflexes due to the slack in intrafusal fibres when muscle is back to its normal length, which reduces background spindle afferent discharge (Burke and Gandevia, 1995; Matthews, 1972; Proske et al., 1993).