Ebook Neurological rehabilitation - Spasticity and contractures in clinical practice and research: Part 2

6
Clinical Management of Spasticity
and Contractures in Spinal Cord Injury
Martin Schubert and Volker Dietz
CONTENTS
6.1Introduction................................................................................................. 136
6.1.1 Epidemiology and Specific Aspects of Spasticity in SCI.......... 137
6.1.2 Spinal Shock, Recovery of Spinal Excitability,
and Development of Spastic Movement Disorder..................... 139
6.1.3 Pattern of Spastic Movement Disorder Depends
on Patho-Anatomy.......................................................................... 141
6.2 Pathophysiology-Based Treatment of Spasticity.................................... 143
6.2.1 Clinical Signs of Spasticity............................................................ 144
6.2.2 Spastic Movement Disorder.......................................................... 144
6.2.3 Therapeutic Consequences............................................................ 145
6.3 Patient Selection and Therapeutic Approach......................................... 147
6.3.1 Indication for Treatment of Spasticity in SCI.............................. 147
6.3.2 Clinical Assessment of Spasticity in SCI..................................... 148
6.3.3 Clinical Presentation and Anatomical Distribution
of Spasticity...................................................................................... 149
6.3.4 Physiological Effects of Training.................................................. 150
6.3.5 The Mainstay of Spasticity Treatment in SCI Is Physical
Therapy............................................................................................. 150
6.3.6 Oral Systemic Anti-Spastic Pharmacotherapy........................... 152
6.3.7 Intrathecal Anti-Spastic Pharmacotherapy................................. 155
6.3.8 Focal Anti-Spastic Pharmacotherapy: Chemodenervation....... 157
6.3.9 Surgical Correction of Contractures............................................ 160
6.3.10 Focal Anti-Spastic Surgical Treatment: Selective Dorsal
Rhizotomy........................................................................................ 161
6.4 The Complex Spastic SCI Patient: Selection of Therapeutic
Approach...................................................................................................... 162

6.4.1 Case 1: Combination Therapies: Oral Systemic and Focal........ 163
6.4.2 Case 2: Combination Therapies: Intrathecal Systemic
and Focal.......................................................................................... 164
References.............................................................................................................. 164

135


136

Neurological Rehabilitation

6.1 Introduction
As in other pathologies involving lesions of the central motor system, spasticity in SCI can be defined as disordered sensorimotor control, resulting from
an upper motor neuron lesion and presenting as intermittent or sustained
involuntary activation of muscles by sensory input. Activation is independent of the type and location of triggering sensory input. It can be touch, pain,
temperature, or proprioceptive stimuli, or it can be mediated by vegetative
stimuli. As in other types of central nervous system (CNS) lesion, spasticity
per se is a pathological condition that is part of a motor syndrome related to
loss of voluntary motor control and related changes in sensory-motor integration and adaptation within the motor system. These changes and adaptations may include adverse as much as beneficial effects for patients’ level of
function and subjective well-being. For instance, it can contribute to muscle
strength and thus function where voluntary strength is lost, thereby supporting stance or gait in incomplete SCI. Hence, spasticity in SCI as much as
in other CNS pathologies may be seen as a compensatory state of a deficit of
sensory-motor control that is usually associated with a lower level of functional CNS organisation. This potentially leads to more disability if negative
effects prevail and balance between voluntary and involuntary activation is
lost. Only in this case is treatment needed. In any case, treatment should be
focused only on these negative effects and should be done with a specific
aim. Such aims can be function, pain control, reducing of care burden, or
prevention of complication such as impending contractures. It must always
involve an interdisciplinary consideration of the patient’s special situation of

impairment. Thus, treatment will usually require that medical staff, patient,
and his/her relatives discuss the treatment aim and agree upon a treatment
concept. This chapter will first deal with the manifestation of spasticity
in SCI and how it can be beneficial or detrimental to function. It will then
describe particular features of SCI spasticity based on spinal syndromes and
their pathophysiology. While there is good understanding of changing excitability of spinal motoneurons below the level of lesion as derived from animal models [1–3], these are not deemed representative of the spastic motor
disorder in human SCI and thus have little meaning in the context of clinical practice. Although there is some experimental work in the human that
supports the notion of changing excitability of infra-lesional spinal motoneurons as a basis for the generation of muscle spasms [4], models derived
from this work rely on several assumptions of analogy with animal models
and have no significance for practical treatment of spasticity in human SCI.
This is mainly due to the fact that the anatomy of the spinal lesion is more
relevant for clinical presentation than modeled excitability changes at the
cellular level. The anatomy of a human spinal lesion results in phenotypes
with implications for functional deficits that have more effect on spasticity
treatment than underlying pathophysiology of presumed neural interaction


Clinical Management of Spasticity and Contractures in Spinal Cord Injury 137

at the spinal segmental level. Therefore, the effects of spasticity in SCI will
be discussed in terms of phenotypes and their implications for function and
need for treatment.
6.1.1 Epidemiology and Specific Aspects of Spasticity in SCI
Spasticity is seen as a major health problem by many patients with SCI [5,6].
Although spasticity can be seen as a compensatory adaptation to the loss of
voluntary motor control, it may also severely limit patients’ mobility when
overshooting and thus can negatively affect independence in activities of
daily living (ADL) and work. Prevalence of spasticity in SCI is reportedly as
frequent as 40–74%, depending on the type of survey and whether external
or self-reported outcomes were drawn upon [1,5–9]. In most surveys, spasticity is rated as the most disabling complication, followed by pain, sexual,

bowel, and bladder dysfunction and pressure ulcers. There is an interrelation
of spasticity, pain, reduced mobility, contractures, and pressure sores [5,6,10].
Many patients report pain as a consequence of spasticity. In fact, spastic and
neuropathic pain can be inseparable in the clinical condition. Independent
of geographic region, the prevalence of secondary health conditions such
as spasticity is known to vary across demographic and SCI characteristics.
Spasticity was more often reported in SCI with incomplete lesions or tetraplegia [5,7,8,10].
SCI as a unique form of CNS damage comes with certain features that are
characteristic to its patho-anatomy. As the lesion is a focused one, severing
the infra-lesional part of the cord from the supralesional CNS, characteristics of SCI will influence the manifestation and the distribution of spasticity.
Neural mechanisms are discussed to be the primary contributors to spasticity following SCI by some authors [9], whereas others emphasise the relevance of mechanisms underlying muscle hypertonia that are unrelated to
increased stretch reflex activity. Intrinsic changes in the muscle tissue itself,
e.g. loss of sarcomeres, histochemical changes, and composition of muscle
fibres, ultrastructure and proportion of extracellular matrix, have been suggested to have a significant impact on spastic hypertonia [11–16]. From a
clinical viewpoint, the original definition by Lance [17] is not sufficient to
understand resulting functional impairment. It is also not helpful in delineating indication for treatment as it does not explain the syndrome of spastic
motor disorder. Clinical signs of spasticity are not related to spastic movement disorder. The functional impairment that follows a central motor lesion
will be influenced and modified by spasticity. However, it is not a direct consequence of the clinical syndrome that was clinically defined by Lance as ‘a
velocity-dependent increase in tonic stretch reflex with exaggerated tendon
jerks, clonus, and spasms, resulting from hyper-excitability of the stretch
reflex’ [2,18]. This is due to several aspects. On the one hand, the definition by
Lance does not capture the signs and symptoms of what is usually referred
to as spastic motor disorder. It does not include the impending secondary


138

Neurological Rehabilitation

changes within muscle and connective tissue leading to contractures as an

unwanted final point of missed treatment.
On the other hand, it overemphasises the significance of the hyperexcitability of the stretch reflex while negating the functional significance of
loss of polysynaptic reflex activity [19]. Spasticity in SCI evolves with time
after lesion. It varies with location of lesion level and other SCI characteristics
such as central cord damage and completeness of the lesion. Clinical aspects
of spasticity are diverse, including muscle hypertonia, flexor or adductor
spasms, clonus, and dyssynergic patterns of contraction. Muscle hypertonia,
an abnormal increase in muscle stiffness, can be regarded as a defining feature of spasticity. Other than exaggerated reflexes, it has both diagnostic and
therapeutic significance [16]. This heterogeneity in clinical presentation cannot be explained by exaggeration of the stretch reflex alone. There is abundance of clinical and experimental neurophysiological work extending on the
suspected mechanisms of spasticity in SCI and the reader is referred to the
respective chapter. However, it should be mentioned that there is controversy
about the putative role of hyper-excitability of spinal motoneurons as a major
cause in the emergence of spinal spasticity. This was put forward based on
the observation of low-frequency invariant spontaneous self-sustained firing
in motor units from 5 out of 15 SCI patients [4]. It was explained as a consequence of altered intrinsic voltage-dependent persistent inward currents
(PICs; e.g., persistent inward calcium currents) [1]. The hypothesis was primarily derived from animal work and then indirectly tested in human SCI
[4]. Under normal circumstances, PICs are assumed to have physiological
roles at the MN level in amplifying synaptic inputs to provide a sustained
excitatory drive that allows motoneurons to fire repetitively following a brief
synaptic excitation. In SCI patients in whom involuntary muscle spasms
could be elicited by various types of afferent stimulation, a self-sustained firing of motoneurons was observed which would last for seconds at unusually
low and regular discharge frequency. Based on several assumptions derived
from animal experiments it was suggested, that this slow spontaneous firing
likely occurs without appreciable synaptic noise and is driven to a substantial
degree by PICs intrinsic to the motoneuron [4]. This would not necessarily
be in contradiction with observations of reduced motor unit action potentials [20] and reduced overall activity of the motor units during functional
movement [12,21–23] as well as a reduction of functional long-latency reflexes
on the one, and enhanced short latency reflex excitability and spontaneous
muscle spasms on the other side [19,24]. However, self-sustained firing of
motoneurons was only observed and described following induced muscle

spasms and not during functional movement. It is unclear whether it could
commonly be observed in chronic spinal injury or if it is only present during
induced spasms. Long-term intramuscular single-motor unit recordings in
the human, which could substantiate the finding, are lacking. It remains to
be determined if there is a relation with functional impairment or if there is a
significant role of the phenomenon in the development of contractures.


Clinical Management of Spasticity and Contractures in Spinal Cord Injury 139

There is more human experimental data supporting the idea that spasticity
involves synaptic mechanisms such as recurrent inhibition [25], reduction in
Ia-reciprocal inhibition [26,27], and reciprocal inhibition of flexor reflex afferents [28]. In summary, changes of motoneuron and interneuron plasticity
are assumed to play a significant role in spinal spasticity, which early after
an SCI are thought related to postsynaptic mechanisms such as receptor upregulation, and later during the recovery phase would be associated primarily with pre-synaptic mechanisms [1,9,29]. However, these changes are not
observed immediately after spinal trauma. They evolve with time, suggesting gradual changes of neural adaptation following SCI.
6.1.2 Spinal Shock, Recovery of Spinal Excitability,
and Development of Spastic Movement Disorder
When describing the natural course of disease following SCI it must be distinguished between pathologies with acute onset and those that result in
slow alteration of the cord, e.g., due to tumor or other etiology with increasing compression. Following an acute onset there will be a phenomenon of
a sudden loss of reflexes and muscle tone, commonly referred to as ‘spinal
shock’. The term was introduced by Hall in 1841, who, in describing the sudden loss and recovery of reflexes, for the first time linked it with the term
‘reflex arc’ [30].
Our present idea is that a flaccid motor paresis is observed immediately
after acute onset of a complete SCI when there are no motor responses to
external stimuli below the level of lesion. During the subsequent days and
weeks, motor reactions to external stimuli and reflex activity gradually reappear in a more or less systematic manner [24]. The phenomenon of spinal
shock remains an issue of debate and controversy. Due to involvement of the
autonomous system in acute SCI, there is some overlap with cardiovascular symptoms, i.e., arterial hypotension and cardiac compensatory response.
The question of duration of spinal shock can be seen as a matter of definition

of the delimiting type of motor reaction or reflex [31]. Depending on what
is chosen as the distinguishing motor criterion, cessation of spinal shock
may be assumed with the appearance of a ‘delayed plantar response’ (DPR),
which occurs within hours after SCI and persists for hours to a few days
[32,33]. If deep tendon reflexes (DTR) are chosen as the criterion, then duration of spinal shock is longer and will comprise several weeks. DTR return
in the majority of patients but the Babinski sign may or may not be present,
which seems to be related to the presence of spasticity [34]. Appearance of
interlimb reflexes indicates late changes reflecting increased polysegmental spinal reflex excitability 6–12 months after SCI [35]. Competitive synapse
growth originating from preserved long descending motor input [36] and
segmental reflex inputs [29] are postulated as underlying the individual outcome and clinical presentation of recovery of voluntary motor control and
spastic motor disorder [35].


140

Neurological Rehabilitation

Complete and incomplete SCI were claimed to be distinguishable by the
extent and duration of spinal shock in several studies lasting only minutes
to hours in ‘slight’ injuries [32,37]. Furthermore, response amplitude to tendon tap and reflex spread to adjacent segments are sensitive indicators of
preserved supraspinal control over lower limb musculature in subjects with
acute SCI and may thus be helpful for prediction of recovery [32]. Conversely,
this would be well in line with the clinical observation of long-lasting flaccidity as an indicator of complete SCI. Within this spectrum of motor responses
and gradually increasing motor activity following spinal shock it is difficult
to distinguish spasticity as a single and clearly defined motor phenomenon.
Spreading reflex activity and clonus is regarded a clinical sign of evolving
spasticity. Muscle hypertonia and polysegmental reflexes may appear as
involuntary contractions and spasms, thus adding to the picture of spastic
motor syndrome of SCI [35]. In the clinical view, the transition from spinal
shock to spasticity is a continuum of an initially gradual increase in motor

excitability [24] with characteristic changes in muscle stiffness, spasms, and
subsequent reduction of short- and increase in long-latency reflex excitability. In contrast to tetraplegic patients, paraplegia resulted in M-wave and
flexor reflex amplitudes that were found to decrease, indicating that spastic motor disorder eventually is not associated with increased excitability of
motoneurons and premotoneuronal network [12,24].
Neurophysiological methods have deepened our understanding of underlying excitability changes in spinal circuits and peripheral nerves during
this transition [20,24,29,38,39]. During spinal shock, the loss of tendon tap
reflexes and flaccid muscle tone is associated with low excitability of spinal motor neurons, as tested by neurographic methods (F-waves) and with
a loss of flexor reflexes, whereas only H-reflexes can be elicited because the
unexcitable intrafusal gamma fibre system is bypassed by direct electrical
stimulation of 1a afferents. Reduced excitability of peripheral mixed nerves
was shown to be based on high threshold stimulus–response relationships
that were apparent from the early phase of spinal shock. This coincided
with depolarisation-like features reaching a peak after 12 and 17 days for
the median and common peroneal nerves, respectively [20,38,40]. Between
Days 68 and 215 after SCI at the end of rehabilitation Boland and coworkers (2011) found that excitability for upper and lower limbs had returned
towards normative values, but not for all parameters. These reductions of
excitability of the peripheral motor axon were described to be paralleled
by the development of spasticity despite reduced excitability of the motor
axon. This supports the notion that spasticity occurs without overactivity
of the motoneurons and their axons. During the transition to spasticity, the
reappearance of tendon tap reflexes and muscle tone can parallel the occurrence of spasms and is associated with the recovery of excitability of spinal motoneurons as indicated by increasing F-wave persistence and flexor
reflex excitability [24] but there is no excess activity of the motor system
causing spasticity. Little change in spinal excitability can be shown after


Clinical Management of Spasticity and Contractures in Spinal Cord Injury 141

this transition phase as the decrease in compound muscle action potentials
(CMAP/M-wave) and reduced flexor reflex amplitude suggest a secondary
degeneration of spinal circuits and motoneurons subsequent to severe spinal

trauma [20,24,41]. Furthermore, flexor reflex excitability depends on the level
of lesion, indicating that spinal interneurons and pre-motoneuronal circuits
may depend on the extent of infra-lesional intact spinal network [24,32]. As
an overall conclusion of these neurophysiological observations during transition from spinal shock to spasticity, it must be emphasised that spasticity in
SCI develops without a net increase in spinal excitability.
6.1.3 Pattern of Spastic Movement Disorder Depends on Patho-Anatomy
Traumatic SCI usually results in a diffuse damage zone of the spinal cord
extending for 2–3 segments, clinically reflected by a ‘zone of partial preservation’. In incomplete SCI, the distribution and extent of segmental damage
is of great relevance for recovery. Contusion injuries inherently represent
the combined damage of both segmental central and peripheral neural
structures [42]. Preserved function of neuronal circuits below the level of
the lesion is the target of rehabilitation training. Spasticity develops only
in this zone. Next to severity and completeness of the injury, clinical spinal
syndromes are relevant as they can show distinct patterns of recovery and
spastic motor disturbance due to specific epidemiology and anatomical distribution of lesion in the spinal cord [43].
The anterior cord syndrome (ACS), due to a flexion injury of the spine,
results in predominant damage of the ventral cord, the segmental ventral
horn cells, and spinothalamic and long motor tracts. This is also possible
when a minor mechanical impact triggers a disturbance of the blood supply
from the anterior spinal artery [44]. In patients with diffuse non-penetrating
spinal injuries, the clinical syndrome is characterised by segmental flaccid
paresis and spastic paresis with disturbance of pain and temperature sensation caudal to the lesion level but sparing of light touch and proprioception,
which are mediated in the dorsal tracts of the cord. Incidence is low, accounting for only 2.7% of all traumatic spinal injuries [45] and less than 1% of all
spinal syndromes [43]. Traumatic ACS as defined by Schneider [46] affects
the anterior two-thirds of the cord and hence involves damage of the lateral corticospinal tracts. This is associated with a poor prognosis and minor
recovery rates of muscle force and poor coordination.
Traumatic central cord syndrome (CCS) is the most common acute incomplete cervical spinal cord injury, accounting for 44% of all spinal syndromes
and for 9% of all SCI in a recent study of 839 spinal cord injuries [43,47].
About 20% of patients with cervical spinal cord injuries present a clinical
CCS [47]. The syndrome is characterised by predominant upper extremity

weakness and clumsy hands, and less severe lower extremity dysfunction
and sensory and bladder dysfunction. Spasticity will be generalised with a
focus on the hands as paresis and loss of motor function is most pronounced


142

Neurological Rehabilitation

here unless lesion level is within the range of the motoneurons supplying the
hand muscles, as this will result in peripheral-type lesion with atrophy and
flaccid paresis. However, most cervical lesions occur at cervical levels C4 to
C6, maximum at C5, while very few affect C7 or C8 segmental levels [43,48],
thus mostly sparing motoneurons of the hand muscles, which are localised
below. CSS represents the oldest age group, with the lowest admission functional level of all SCI clinical syndromes, which is a cofactor in determining
relatively poor recovery of hand function in this group, despite its favorable outcome compared to traumatic incomplete cervical SCI in general [43],
which is in the range of the group of Brown-Sequard [49]. Hand spasticity
in these patients can add to their functional impairment in activities of daily
life due to loss of manual dexterity. However, walking ability can also be
severely impaired by spasticity of the trunk and legs. CCS was originally
thought to result from post-traumatic centro-medullary hemorrhage and
edema [50], or from a Wallerian degeneration, as a consequence of spinal
cord compression in a narrowed canal [47].
The central focus of spinal damage in combination with the special
somatotopic organisation of the corticospinal tract, where motor tracts for
the upper are localised more centrally than those for the lower extremities,
were assumed to be responsible for the predominance of motor deficits in
the hands in CSS. However, more recent anatomical analysis and primate
animal studies suggest that the syndrome is due to the specific effects of a
cervical spinal lesion on direct corticomotor (pyramidal) tracts given their

significant role in manual motor control [51]. This would be in line with
the seminal findings of these direct cortico-motoneuronal projections by
Bernhard and Bohm [52] and with these authors’ appreciation and consideration of this anatomical feature, which is unique in primates and humans.
A loss of the capacity for ‘fractionation’ of movements and control of small
groups of muscles in a highly selective manner [53] is as much characteristic of CCS as an impairment of the acquisition of new motor skills [54].
Therefore, when considering the significance of direct cortico-motoneuronal
control in human manual dexterity [51], CSS may be considered a prototypical condition where spinal cervical lesion inflicts damage predominantly on
pyramidal tract axons affecting fine motor control and coordination of the
hand. Loss of fine motor control in general and, hence, particularly in the
condition of CSS is associated with spastic motor disorder, which can lead
to contracture and pain, predominantly in the upper extremity. This mostly
concerns the flexor muscles of the hands.
A hemisection of the cord leads to Brown-Séquard Syndrome (BSS),
which was first described in 1851 by the neurologist Charles Edouard
Brown-Séquard [55] as ipsilateral ataxia and spastic paresis due to proprioceptive and motor loss in association with contralateral loss of pain
and temperature sensation below the level of lesion. A surgical unilateral
lesion dividing most of the ipsilateral tracts of the spinal cord resulted in
complete flaccid paresis of the ipsilateral limbs only for a few hours, after


Clinical Management of Spasticity and Contractures in Spinal Cord Injury 143

which voluntary movements began to reappear [56]. Within days after such
a sharp lesion, patients were able to exert slow digital movements, and
walking ability was attained within 2 weeks. Slow and feeble manual function recovered within less than 3 weeks of the operation. This indicates
that recovery and redundancy in corticospinal control is strong in human
SCI. However, this syndrome is rare in traumatic SCI and its recovery is
generally less favorable than in the cases with a sharp penetrating spinal
lesion, as described by Nathan, indicating that there must be more extensive and diffuse lesion of spinal tracts in lateralised traumatic SCI [57].
Although BSS-like syndromes with more or less lateralisation of lesion are

relatively rare in Europe and account for less than 4% of all traumatic SCI
[43], they are nevertheless relevant as prognosis is known to be most favorable among incomplete traumatic SCI [43,57,58], particularly with regard
to ambulation. Physiologically, recovery occurs in a rather characteristic
order, with proximal extensors prior to distal flexors on the more affected
side and vice versa on the less affected side [58]). This is attributed to the
unilateral (distal flexors) and bilateral (proximal extensors) distribution of
preserved fibres and their recovery due to sprouting and formation of collaterals. The recovery is most likely owed to lumbar midline crossing fibres
[59,60]. Spasticity usually is present, but does not pose a problem in these
patients.
Conus medullaris syndromes amount to 1.7% and posterior cord syndrome to less than 1% in the analysis of McKinley and coworkers [43]. Data
on these groups are sparse. In general, spinal syndromes tend to need shorter
rehabilitation length of stay, indicating that sufficient functional outcome is
reached after shorter duration of rehabilitation, which is likely secondary
to an in-complete pattern of lesion and high proportion of preserved spinal
nerve fibres [43]. Spasticity usually only occurs in the plantar-flexors and
digital muscles where there is an epi-conus lesion leaving intact ventral horn
motoneuron cells that are disconnected from supraspinal input.

6.2 Pathophysiology-Based Treatment of Spasticity
Spasticity even today is frequently thought to be reflected in an ‘extraactivity’ in limb muscles mediated by exaggerated reflexes leading to muscle
overactivity. Also, most articles in this volume are focused on these phenomena. The consequence of this thinking is that spasticity should be treated
by attenuating reflex and muscle activity by antispastic drugs or botulinum
toxin injections. However, for over 40 years convincing evidence has been
available indicating that these assumptions hold only partially for ‘clinical
spasticity’ but not for spastic movement disorder, which hampers the patient
(for review [61]).


144

Neurological Rehabilitation

In contrast to clinical signs of spasticity, it is characterised by a reduced
limb muscle activation. According to the studies on spastic movement
disorder, secondary to a CNS lesion, alterations of mechanical muscle
fibre properties occur in association with low tonic muscle activity, which
allows the development of spastic muscle activity to compensate the
reduced dynamic muscle activation during functional movements after,
e.g., a stroke. This enables the patient, for example, to support the body
during stepping. The consequence of this compensatory mechanism in
mobile patients is that anti-spastic drugs can accentuate paresis. In the
following paragraphs, we will discuss the multiple aspects of evidence in
more detail.
6.2.1 Clinical Signs of Spasticity
The diagnosis of a spastic paresis is based on the examination of tendon
tap reflexes and muscle stiffness in the passive subject. Early after an acute
damage of the CNS, tendon tap reflexes are exaggerated, but muscle stiffness
develops only after some weeks. When stretching a limb muscle of a spastic patient (Ashworth Test) during the clinical examination a tonic muscle,
activation occurs in this muscle, leading to an increased resistance [62]. This
observation has led to the assumption that exaggerated reflexes result in an
increased muscle activity and, consequently, are responsible for the movement disorder. However, electrophysiological investigations on the neuronal
adaptations after a complete spinal cord injury indicate a divergent course of
increasing clinical signs of spasticity but decreasing or stable values of their
potential neuronal correlates (M-wave, F-wave, H-reflex, and flexor reflex)
[24]. Consequently, non-neuronal mechanisms were assumed to contribute
to spastic muscle stiffness. In addition, according to all investigations of
natural, complex movements in patients with spasticity, the assumption of a
relevant ‘extra-activity’ contributing to spastic muscle stiffness could not be
confirmed [19].
6.2.2 Spastic Movement Disorder

For a patient with spasticity, the impaired performance of hand or leg/
stepping movements and their treatment are of importance, not the clinical signs found during examination. During active movements such as gait
a low amplitude, tonic activation of upper and lower limb muscles can be
observed, i.e., a normal modulation of EMG activity is lacking while a normal timing of muscle activity is largely preserved [12,63]. The reduction of
limb muscle activity is suggested to be due to a diminished excitatory drive
from supraspinal centers and an attenuated activity of certain polysynaptic
(or long-latency) reflexes [64,65]. Polysynaptic reflexes are known to modulate limb muscle activity [64] and thereby adapt the movement pattern to the
environmental requirements.


Clinical Management of Spasticity and Contractures in Spinal Cord Injury 145

In contrast, short latency reflexes neither in healthy subjects nor in
patients with spasticity contribute significantly to muscle activity during
natural movements [19]. These observations indicate that the muscle activity required during movement performance (e.g., to support the body during the stance phase of stepping) develops on a lower level of organisation
after a CNS damage [19,61,66]. Consequently, the muscle tone required is not
achieved by a modulated muscle activation as it is the case in healthy subjects. Instead, muscle hypertonus develops with the stretching of the tonically activated muscle. This represents a more simple mode of muscle tone
generation, which is also based on structural alterations of a muscle secondary to a CNS lesion, i.e., a loss of sarcomeres [66], muscle fibre changes and
increase of structurally deteriorated extracellular matrix [14–16]. Increased
passive tension in the muscle is unrelated to stretch reflex activation. At the
single-fibre level, elevated passive tension was found in muscle cells expressing fast myosin heavy chain isoforms, especially MyHC-IIx, but not in those
expressing slow MyHC. Type IIx fibres were present in higher-than-normal
proportions in spastic muscles, whereas type I fibres were proportionately
reduced [16]. This is equivalent to an alteration of the contractile properties
toward tonic muscle characteristics. According to these authors, ultrastructural changes of the extracellular matrix such as expanded connective tissue, but also decreased mitochondrial volume fraction and appearance of
intracellular amorphous material, suggest that the global passive muscle
stiffening in SCI spasticity is caused by structural and functional adaptations outside and inside the muscle cells, which alter their passive mechanical properties. This change compensates in part for the loss of neurogenic
muscle activation and allows, for example, for support of the body during
the stance phase of stepping. However, the performance of quick/fast movements becomes impossible by this mode of regulation of muscle stiffness.
Muscle spasms do not play a role in this. Patients with spasticity do not only

suffer from an impaired motor output but a defective control and processing
of afferent signals contribute to the movement performance [65].
Thus, in patients with spasticity, in comparison with healthy subjects,
muscle activity is enhanced in the passive state, i.e., during the clinical examination, but is reduced during active natural movements. The spastic signs
observed during the clinical examination can therefore hardly be translated
to the movement disorder. Clinically, spastic signs are more pronounced in
damage of the spinal cord compared to a cerebral lesion. However, from a
pathophysiological point of view there exist only quantitative but no qualitative differences.
6.2.3 Therapeutic Consequences
Exaggerated reflexes do little to contribute to the movement disorder that
impairs the patient. Nevertheless, most anti-spastic drugs are directed to
reduce the activity of short-latency reflexes mediated by group Ia fibres in


146

Neurological Rehabilitation

order to reduce muscle stiffness. However, mobile patients require spastic
muscle stiffness to support their body during stepping to compensate for
paresis [61]. Therefore, anti-spastic drugs can accentuate paresis and consequently can lead to a worsening of function. Similarly, some authors argue
that botulinum toxin type A is assumed to result in a largely cosmetic effect
on spastic signs without functional improvement [67,68], although this toxin
might reduce the activity of the intrafusal fibres [69,70]. Intrathecal baclofen
might also reduce hyperactive reflexes without producing significant weakness [71–73]. In conclusion, therapeutic interventions in patients with spastic
paresis due to an incomplete SCI should be focused on the training, relearning, and activation of residual motor function [74,75], and the prevention
of secondary complications, such as muscle contractures [76]. Anti-spastic
drug therapy might predominantly benefit immobilised patients by reducing muscle stiffness and relieving muscle spasms [77], which might in turn
improve nursing care for these patients. In cases where function is hampered by a focal imbalance of specific muscle groups resulting in movement
impairment or contracture, focal botulinum toxin is known to be effective

in improving pain, helping to avoid or to reduce contractures, and facilitating function. Its action is by a weakening and relaxation of muscle activity
resulting in a biomechanical change in the muscle’s function. It makes the
muscle amenable to stretching and lengthening in order to restore to some
extent the interaction of antagonists. Thus, in addition, the weakening of the
agonist allows to some extent a strengthening of the antagonist muscles and
thereby it is possible to restore some of the disturbed antagonistic balance
[78]. This is independent of mobility of the patient but will require at least
some mobility of the affected limb when targeting functional improvement.
In contrast, mobile patients can benefit from a functional arm and leg
(locomotor-) training, which is associated with a recovery of function [19,61].
In animal experiments it could be shown that afferent signals induced by the
functional training to spinal cord neurons below the lesion lead to a directed
neuroplasticity [79] that is associated with a physiological mode of limb muscle activation. In contrast, according to this study, a lack of training of natural
movements leads to a chaotic sprouting associated with a neuronal dysfunction, which might hamper a successful regeneration in the future in chronic
SCI subjects [80]. The clinical consequence of a functional training in mobile
patients is that with the improvement of function during the course of training less spastic muscle stiffness is required for movement performance, i.e.,
a new equilibrium between improved mobility and less pronounced signs of
spasticity becomes established [61].
As a consequence it follows that, in mobile patients, anti-spastic medication can impede recovery of natural movements, as the performance of
natural movements requires some spastic muscle stiffness for compensation
of the paresis, i.e., lack of sufficient muscle activation [81]. Robotic devices
can support this repetitive training. They allow longer training times and
can provide useful feedback information to the patient about the course of


Clinical Management of Spasticity and Contractures in Spinal Cord Injury 147

functional recovery [82]. In immobilised patients a neuronal dysfunction
develops about 1 year after injury [21,83] likely as a consequence of the loss
of afferent feedback signals due to the immobility. This is reflected in rodent

experiments by an undirected sprouting of tract fibres below the level of
lesion [79].

6.3 Patient Selection and Therapeutic Approach
A consideration of treatment of spasticity is made when the patient or the
attending medical team observes persistent or pending signs of impairment
or harm associated with spasticity. As spasticity evolves with time after SCI
(due to spinal shock in severely affected patients), this is more likely to occur
some weeks after the injury during rehabilitation. It is important to follow
a strategy ruling out possible external triggers and after analyzing the exact
circumstance of the phenomenon before initiating a treatment. Thus, it is
important to obtain the view of the other members in the medical team and
inquire about the observations of the patient and their relatives prior to the
decision to treat. Management of these patients is teamwork, as is the entire
rehabilitation of SCI. Initial questions will pertain to the level of independence and mobility of the patient. While there are exceptions, ambulatory
function mostly may preclude or limit treatment approaches with intrathecal
application of baclofen, which is mostly reserved for immobilised patients
with severe incapacitating spasticity leading to contractures.
6.3.1 Indication for Treatment of Spasticity in SCI
Despite the complex theoretical and pathophysiological knowledge of underlying mechanisms associated with alteration of stiffness and reflex function,
the management of spasticity in SCI is to a large extent empirical. An indication for treatment of spasticity in SCI exists when it may cause harm and
interference with function, nursing, or subjective well-being [78]. This may
be expressed by the patient or by the nursing staff and treating therapist
and physician [84]. A consensus should be reached as to the reason to treat
and treatment aim [78,84]. The most common treatment goals in spastic SCI
are enhancement of mobility and speed, increase of endurance and speed of
ambulation or wheelchair propulsion, improvement of transfers, improvement of reaching, grasping, grooming and dressing, relief from pain, and
painful muscle spasms, improvement of tolerance to wear splints and orthosis, which in turn will be needed to improve mobilisation of limbs and secure
therapy effects aimed at prevention of contractures, prevention of contractures, promotion of hygiene, improve positioning, and facilitate mobilisation
and other therapies [78].

148

Neurological Rehabilitation

Depending on distribution of spastic symptoms and their interference
with function, mobility for nursing aims, or subjective well-being of the
patient, a treatment regimen can be chosen [78,84]. Management of spasticity is always a multidisciplinary and polymodal process, which includes
physical and pharmacological measures. It has to be considered that the
effects of spasticity might not always be negative. Spasticity can stabilise
weakened legs, allowing a patient to stand or transfer and have improved
mobility. Spasticity can also be a functionally helpful factor by being protective against skeletal muscle atrophy, decreasing the incidence of fracture.
Moreover, spasticity has been reported to increase glucose uptake and will
improve metabolism, thereby reducing the risk for diabetes in SCI [85] as
well as augmenting cardiovascular function and energy consumption. The
goal of treatment of spasticity must therefore consider the balance of functional benefits from spasticity and its adverse effects in allowing and facilitating motor function and nursing in immobilised patients.
6.3.2 Clinical Assessment of Spasticity in SCI
Prior to any initiation of treatment, it is essential to have a thorough description of the extent and degree of the spasticity. Furthermore, the patient’s
day-to-day functioning should be known. Spasticity can prevent simple
maneuvers essential in daily life, such as transfer and the placing of hands
and arms to control an electric wheelchair, rendering manual hygiene or catheterisation difficult. Personal accounts of the patient as well as information
from those who know the patient should be obtained. This is particularly relevant when planning treatment of focal spasticity with chemo-denervation,
as it must be determined after first injection of botulinum b toxin whether
the dosage and pattern of application is optimal. When evaluating and discussing treatment options, a clear goal should be determined for what is
to be achieved by the treatment. In the clinical examination, it is important to assess the range of active and passive movements as well as painful
limitations of movement or abnormal limb positions. While not functionally relevant, as it is not strongly related to loss of function, the most widely
used assessment scales are the Ashworth Scale and the Modified Ashworth
Scale [86,87]. It is therefore not recommendable to assess treatment effects,
except in testing response to intrathecal baclofen (see below). Other scales

come with the same limitations [88–90] and are therefore of limited clinical
value. Assessment can be done with the aid of video clips from before and
after treatment [84]. Electromyography (EMG) can be useful to identify and
inject spastic muscles in focal treatment by chemo-denervation. However,
EMG cannot be used to assess degree of spasticity. Individual patient history, with an emphasis on functional limitations of specific activities in daily
life, are more helpful to determine, if treatment is effective and satisfactory.
There are established and validated scales and scores in SCI to assess and
quantify activities of daily life (ADL), e.g., the Spinal Cord Independence


Clinical Management of Spasticity and Contractures in Spinal Cord Injury 149

Measure [91–93]. SCIM and other scores cover the main ADL domains relevant to SCI, such as mobility (6 minute walking test, 10 m walking test,
walking index in SCI: WISCI) [94,95], self-care (SCIM), unilateral hand function (Graded Redefined Assessment of Strength, Sensibility, and Prehension:
GRASSP) [96,97], and bladder and bowel management (SCIM) [98]. Most of
these scores have been validated and shown to be responsive to change and
can serve as a tool to evaluate effects of anti-spastic treatment if function is
a treatment aim [93,99]. In any case, it is recommendable to include patientreported outcomes relating to quality of life and participation when defining
treatment outcomes in any individual case and in clinical trials. Assessment
prior to and after treatment should then include both patient-reported and
externally rated functions in the activity of daily life. These are functionally
relevant and can contribute to a patient’s well-being, while scores for the
rating of clinical spasticity are not suited, and will likely not contribute to, a
patient’s benefit from treatment.
6.3.3 Clinical Presentation and Anatomical Distribution of Spasticity
Patho-anatomical distribution and severity of spinal lesion, among other factors, determine localisation of spasticity. Spastic symptoms may be more or
less focal or regional, e.g., most prominent in the upper extremity in a central cord syndrome or pronounced in the legs in a thoracic complete spinal
lesion. Generalised spasticity may affect the trunk and abdominal muscles,
leading to pain or respiratory constraints. Spasticitiy of the upper extremity after SCI typically presents with shoulder adduction and inward rotation, elbow, wrist, and finger and thumb flexion, and pronation. Typically,
patients’ hands tend to be fixed with closed fists, resulting in an impairment

of reaching, grasping, and releasing. Spastic hypertonus of the pelvic striate
muscles can impede micturition and defecation. Distribution and localisation of symptoms will guide the choice and form of application of anti-spastic
treatment. A sudden increase in spasticity should prompt the attending physician to screen for underlying pathology that can be completely independent of the spastic motor disorder. The first step in the management of all
problematic spasticity is to identify, address, and treat any remediable causes
and factors [100] such as an over-filled bladder, obstipation, acute infections,
syringomyelia, or bone fractures may substantially influence the degree of,
or suddenly initiate, spasticity and must be determined [84]. An assessment
of the clinical and functional consequences for the patient is decisive before
management. If such measures are ineffective then it is appropriate to pursue or increase medical treatment until a therapeutic response is obtained. It
is important to notice and attribute sudden changes in spasticity especially
in SCI because infection of the urinary tract, fever, constipation, skin lesions,
and local bone or joint injuries may not present in the usual way and go
unnoticed below the level of lesion. In fact, increase of spasticity may be the
leading and only symptom. Consequently, worsening of spastic symptoms


150

Neurological Rehabilitation

should prompt search for the most common and likely ailments of SCI
and spastic symptoms may immediately be alleviated by appropriate (and
causal) treatment of bladder infection or skin sores. Therefore, a thorough
check of patient history (e.g., for recent falls or injuries) as well as a thorough
examination should be initiated in cases of sudden change of spasticity. If
triggers are found their causal treatment may be prior and effective to also
treat harmful spasticity.
6.3.4 Physiological Effects of Training
A slowing of peak velocity of plantar-flexor muscles was shown to be related
to loss of strength and spasticity in SCI patients [101]. Spasticity and injury

level determine the pattern of abnormality in gait after spinal cord injury
[102]. Spasticity and paresis may thus be seen as directly related to the extent
of functional motor impairment. There are few systematic studies on this
issue [103–105] and recent work shows that functional improvement can be
induced by different types of functional motor training while simultaneously affecting volitional control and spasticity [106]. Endurance and precision training were shown to facilitate descending excitatory as well as
spinal inhibitory networks in patients with incomplete SCI in parallel with
improvement of walking function and reduction of the cutaneo-muscular
reflex excitability. The latter involves excitatory and spinal inhibitory components. Training-induced parallel increase of volitional control and spinal
inhibitory components of the cutaneo-motor reflex suggests that spared
descending pathways originating from the motor cortex can be strengthened by the intervention, hence increase of motor control will parallel a
decreased susceptibility for involuntary muscle spasms [106,107]. Thus, various types of functional motor training and physiotherapy may, in addition
to strengthening the descending excitation of the spinal cord, also increase
the strength of inhibitory spinal networks activated by both descending and
peripheral afferent pathways. These neurophysiological changes may then
lead to improvement of volitional control of movement as well as reductions
in involuntary muscle spasticity, such as reflected in the reduced spasm-like
cutaneo-muscular reflex.
6.3.5 The Mainstay of Spasticity Treatment in SCI Is Physical Therapy
As was argued earlier, all patients with spasticity should be urged to exercise. If this is not sufficiently effective, the patient should have physiotherapy
with guidelines for exercises that counteract the spasticity. Physical therapy does not have to be costly and can be performed by any caregiver who
supports and guides a patient in his activity. For an immobile patient who
is able to move with help, regular transfers and mobilisation twice a day
may be critical to prevent contractures. For someone who can barely stand
or walk, such a help for active mobilisation twice a day may be crucial for


Clinical Management of Spasticity and Contractures in Spinal Cord Injury 151

him in order not to lose his remaining capacities. We have often seen spastic patients who, once bed-ridden, never regained their former ambulatory
capacity without any other cause than transient immobilisation. The most

important factor is regular activation in short bouts that do not exhaust the
patient. Systematically increasing training intensity in order to keep stress
levels low is likely recommendable as a basic principle of physical therapy
and may improve any outcome [108].
If no treatable causes triggering spasticity can be found, the team should
decide whether focal or general signs of spasticity prevail and require treatment. In focusing on functional deficits, the first level of treatment and the
basis for any further escalation should be physical therapy. This recommendation is based on the notion of the ambivalent role of spasticity in central
paresis and impaired motor control. In a systematic analysis of the disturbed
motor control following SCI [19,109] it was suggested that replacing lost patterned activation of the spinal cord by activating synaptic inputs via assisted
movements and/or electrical stimulation may help to recover lost spinal
inhibition, thus leading to a reduction of uncontrolled activation of the spinal cord to improve its function [109]. Increasing the excitation of the spinal
cord with spared descending and/or peripheral inputs by facilitating movement, instead of suppressing it pharmacologically is therefore the primarily
suggested approach to improve residual motor function and manage spasticity after SCI. Any treatment of spasticity will then be a combination of physiotherapy passively mobilising the spastic limb or body part and increasing
active movement within the limits of residual motor function. Physiotherapy
can be administered whether the patient is mobile or immobilised. Therapy
can be tailored to the patient’s needs and capabilities. It is associated with
the welcome effect of personal attention. Water therapy in itself can help to
reduce muscle stiffness and will, as a side effect, reduce tension-inducing
load as it eliminates gravity and thereby reduces defensive hypertonus and
alleviates mobilisation.
Recently, treatment concepts such as those described by Bobath and Vojta
(for review see [110]) have not been pursued very rigorously. While primarily
used in pediatric facilities in the past for treatment of spastic cerebral palsy,
they are not common concepts in SCI treatment. They are worth mentioning as systematic approach with the scheme to activate complex stereotyped
movement patterns that are believed to reside in the network of the spinal
cord (Vojta) or to inhibit spastic symptoms in flexor muscles of the upper and
extensors in the lower extremities (Bobath). However, there are no validated
studies to support this notion.
Locomotor training (LT) has become a standard of treatment of leg muscle function for ambulation and stance balance control in SCI [110–112]. The
observation that LT can ameliorate spasticity is based on observations made

in cats with complete spinal lesions [113]. LT on a treadmill is combined with
bodyweight support, reducing gravitational forces by 20–50% by means
of mechanical support by an overhead harness. As subjects walk on the


152

Neurological Rehabilitation

treadmill with a reduced load on their lower extremities, coordinated stepping movements and patterned muscle activation can be facilitated by the
moving treadmill. In a recent review evaluating the current LT approaches
for gait rehabilitation in a total of 384 individuals with incomplete SCI, it was
shown that evidence on the effectiveness of locomotor therapy is still limited
[114]. While all approaches showed some potential for improvement of ambulatory capacity there is no superior method and effects were limited. Main
effects in the included studies were shown for gait velocity and distance,
hence a functional improvement which may be assumed to be associated
with an indirect improvement of muscle tone and its regulation. However,
only a subgroup of studies included measures of spasticity and this outcome
was neither systematically analyzed nor reported in the review.
6.3.6 Oral Systemic Anti-Spastic Pharmacotherapy
If necessary, generalised spasticity can be treated with oral medication.
Several drugs with antispastic effect are available with various mechanisms
of action. Baclofen, tizanidine, benzodiazepines, gabapentin, clonidine, and
cannabinoids are centrally acting drugs. Gabapentin, clonidine, and cannabinoids are not officially approved for the treatment of spasticity in SCI.
They are well-established drugs with known risk profiles and some lowlevel evidence for antispastic activity and may therefore be considered as
second- or third-line treatment options in SCI spasticity on an individual
basis. Dantrolene and botulinum toxin type A have peripheral action. The
latter will be discussed in a separate paragraph. A synopsis of available oral
pharmaceutics, their main effect of action and metabolism, and side effects
is given in Table 6.1. Some of these drugs were especially developed to treat

spasticity after SCI. For instance, baclofen was first introduced in 1964 for
this use. Baclofen is structurally similar to g-aminobutyric acid (GABA). It
binds to GABA-B receptors in the brainstem and dorsal horn of the spinal
cord. By suppressing the release of excitatory neurotransmitters involved
in monosynaptic and polysynaptic reflexes, it is assumed to reduce muscle
stiffness and spasms [115].
Treatment effects of oral anti-spastic drugs are not well-documented. Most
of the studies upon which anti-spastic pharmacological treatment has so far
been based date from the 1970s, 1980s, and 1990s of the last century. While
still being cited in modern work, these early studies lack modern standards
of good scientific and clinical practice and the paucity of newer work may, at
best, indicate that there is little interest to build a better base of evidence in
pharmacological treatment of spasticity in SCI. New literature often appears
in the form of reviews of the same poorly controlled original studies. Such
systematic reviews repeatedly show that there is insufficient evidence to
assist clinicians in a rational approach to anti-spastic treatment for SCI
[85,116]. These reviews, as well as several others, showed that study quality is
low. Thus, therapy is mostly empirically guided, and randomised controlled


24–48

2 × 25–5 × 50

2 × 0.075–3 × 0.15
3 × 100–3 × 800

THC and CBD* (Sativex)

Dantrolene (Dantamacrin)

Clonidine (Catapresan)
Gabapentine (Neurontin)

Alpha 2
GABA (indirect)

Formatio reticularis,
polysynaptic
reflexes
ECS (CB1, CB2
receptors)
Striated muscle

GABA-A (spinal
and cerebral)
Alpha 2

GABA-A (spinal)

Mechanism

Hepatic
Renal

Hepatic

Hepatic

Hepatic

Hepatic

Hepatic

Hepatic

Metabolism

Dizziness, sleepiness,
nausea, mood disorder
Hepatopathia, sleepiness,
nausea, weakness,
Hypotonus, bradycardia
Dizziness, sleepiness

Sleepiness, nausea,
weakness, hypotonus,
respiratory depression (IT)
Sleepiness, nausea,
weakness, hypotonus
Sleepiness, nausea,
weakness, hypotonus,
dry mouth
(Rare) gastro-intestinal
symptoms, hypotonus

Side Effects

Second line due to side

effects
Second line
Second line, for spastic
pain

Aerosol

Acts primarily on trunk
muscles

Slow dose increase!

Caution: withdrawal;
intrathecal dose 1/100 to
1/1000 of oral dose
Caution: withdrawal

Specials

a

Baclofen, clonazepam, tizanidine, and tolperisone are first-choice drugs; THC and CBD, dantrolene, clonidine, and gabapentine are second-choice
drugs.
* Oral spray, mix of: THC = Delta-9-Tetrahydrocannabinol; CBD = Cannabidiol, maximum dosage 12 × 100 ul.

3 × 150

3 × 2–4 × 6

Tizanidine (Sirdalud)

Tolperisone (Mydocalm)

0.5–3 × 2

3 × 5–3 × 30

Baclofen (Lioresal)

Clonazepam (Rivotril)

Dosage (mg)

Drug (Trade Name)

Common Oral Antispastic Medicationa

TABLE 6.1

Clinical Management of Spasticity and Contractures in Spinal Cord Injury 153


154

Neurological Rehabilitation

trials (RCT) are rare and done only in very limited numbers of patients. A
large systematic review with specific focus on SCI [116] returned a yield of
262 references, of which only eight met the inclusion criteria of cross-over
RCT, with 100 patients (80 males) enrolled in these 8 cross-over studies, 14 of

which were spinal forms of multiple sclerosis. Three of these studies tested
efficacy of intrathecal application of baclofen (ITB; see paragraph below). One
parallel multi-center trial compared tizanidine with placebo [117] among
118 patients (104 males), all of whom had an SCI (traumatic etiology in 108
patients) at cervical and thoracic level, proving efficacy without additional
loss of strength. However, there was no improvement of ADL [117,118].
Outcomes in most studies were measured in terms of spasticity scales
(MAS or other), while only one study assessed the performance of activities
of daily life, showing no improvement [117,118]. The poor quality of included
studies, and the marked differences in study designs, outcome assessments,
and methods of reporting, did not allow for the performance of a quantitative combination (meta-analysis) of the results. The same applies to other
similar reviews [85,116].
While efficacy is low, adverse drug reactions were reported to be common.
Within the scope of these reviews, only a couple of studies were of direct
comparisons of antispastic drugs (e.g., tizanidine vs diazepam or baclofen),
where no significant differences were found clinically [85,116], whereas differential effects on flexor and extensor leg muscles were observed in a direct
comparison of baclofen and tizanidine, while neither drug caused weakness
at low dosage [119]. However, in the latter study on 10 chronic SCI patients, no
clinical or functional data were presented. The general methodological quality of most studies was poor according to the systematic review by Taricco
et al. [116] was poor, impeding meta-analysis or firm conclusions regarding
the clinical management of spasticity. Poor efficacy of anti-spastic drugs on
muscle hypertonus was attributed to the fact that most anti-spastic drugs
reduce reflex activity. In contrast, as pointed out earlier, recent pathophysiologic evidence has suggested that exaggerated reflexes contribute little to
spastic muscle hypertonia [61]. In the majority of mobile patients, impairment of functional movements is clinically more relevant than impairment
of muscle tone. Functional movements were only assessed in half of the trials. Daily living activities and the overall patients’ status were also rarely
assessed, which contrasts with the therapeutic objective of routine clinical
practice. In conclusion, for various reasons, there was not enough evidence
from available clinical trials to assess, and compare, the effects of drugs commonly used to relieve spasticity after spinal cord injury. Hence, the overall
perception on oral anti-spastic treatment in SCI is today one of obfuscation,
best expressed by the following quote from one recent review: ‘published

reports depict a […] gloomy panorama on the treatment of chronic spasticity
by oral route’ [85].
Based on empirical recommendations, a combination of pharmacological agents at low dosage is assumed to help reduce side effects while


Clinical Management of Spasticity and Contractures in Spinal Cord Injury 155

increasing efficiency. The choice of the drug or combination thereof will
be made based upon the individual response of the patient, age group,
prior experience, price, and profile of side effects and potential interaction
with other medication (e.g., tizanidine must not be combined with gyrase
inhibitors such as ciprofloxazine, as this would cause relevant interaction with the toxic effects of the anti-spastic drug). Any of these systemic
treatments should be amended with focal physiotherapy, but can also be
combined with focal chemodenervation (botulinum toxin or phenol, see
below). This is more relevant, as adherence to anti-spastic medication is
problematic, hence dosage should be kept minimal when combined with
non-pharmacological treatment. In a large study of 2840 subjects with various types of central motor syndrome (including stroke, spinal cord injury,
traumatic brain injury, cerebral palsy, and multiple sclerosis), adherence
to anti-spastic medication was at best 50% of treatment periods [120]. This
may indicate an unmet need for better anti-spastic medication and better
guidance with treatment.
In a recent systematic pharmacological approach, efficacy of spasticity
treatment with tetrahydrocannabinol (THC) was assessed in a placebocontrolled trial in 25 SCI patients [121]. A major reason for drop-out was the
increase of pain and psychological side effects. The latter should be reduced
when using a combination of THC and cannabidiol, a partial antagonist that
is now commercially available as an aerosol for the treatment of spasticity in
multiple sclerosis [122,123]. However, when compared to established drugs,
the cost of this preparation is high.
6.3.7 Intrathecal Anti-Spastic Pharmacotherapy
The effective treatment of generalised spasticity is achieved by intrathecal

application of baclofen (ITB), first introduced by Penn and Kroin in 1984 [89].
Studies on outcome measures such as the Ashworth Scale and spasm score
as well as studies assessing quality of life have suggested the superiority of
ITB over oral baclofen [124,125]. Despite a lack of trials directly comparing
oral administration of baclofen and ITB, it is commonly agreed that ITB is
indicated when spasticity continues to produce a clinical disability, despite
trials of high dosages of oral treatments in patients who have functional
goals and/or significant pain and disability. ITB allows for flexible dosing
patterns to suit an individual patient’s lifestyle [100]. Thus, whenever generalised spasticity cannot be adequately controlled with oral medication,
i.e., due to insufficient effects on muscle hypertonus despite maximum dosage or due to intolerable side effects at sufficient dosage with a combination of oral drugs, reversion to ITB should be considered. Severe sequelae of
spasticity, such as contractures or progressive neurogenic scoliosis, may be
among the conditions that should prompt consideration of ITB where applicable. Despite the considerable cost of the device and the effort required to
test efficacy for each patient prior to implantation, this route of application


156

Neurological Rehabilitation

has clear advantages over oral administration. Efficacy is empirically superior to that of oral therapy and can be achieved with little or no side effects.
The effect of ITB is estimated to be 100 times greater than with oral administration. Due to the much-enhanced pharmacological effect, patients do
not experience drowsiness or dizziness, while muscle hypertonus can be
effectively reduced in the lower extremities and in the trunk. Patients with
high degree of functional impairment due to immobilisation are most likely
to benefit from this therapy. There is some limitation of efficacy in the arms
and shoulders because with ITB the concentration of baclofen diminishes
in the cranial direction [126]. Furthermore, effects in the upper extremity
will be weaker due to the recommended position of the intrathecal catheter
no more cranially than the level of Th1, to avoid central effects of the drug,
such as depression of respiratory function and vigilance. It is assumed that

a more cranial position of the catheter may lead to toxic concentrations of
the drug at the brain stem level and thereby significantly increase the risk
of critical side effects associated with depression of respiratory and rhythmregulating midbrain centers. Furthermore, positioning of the catheter tip
should be optimised with respect to the main focus of spasticity. If it is in
both the upper and lower extremities then one should attempt to place the
catheter tip as high as T1; if the spasticity only affects the lower extremities,
it can be placed between T6 and T10 [100].
It is recommended to evaluate the patients and their caretakers to determine whether they and caregiving teams meet the demands required to
ascertain pump management and maintenance. It must be explained to the
patient and their caregivers that it is crucial to make follow-up appointments
to keep track of effects and adverse events. Relative contraindications are
anticoagulant therapy with coagulation disorders, anatomic abnormality
of the spine, and localised or systemic infection. It is generally agreed to
only implant ITB pumps in non-ambulatory patients. This is due to the fact
that the effect of baclofen on muscle hypertonus is non-specific and will also
pertain to volitional muscle activation. In fact, in spastic patients it may be
impossible to distinguish spastic tone from volitional strength directed to
main posture. Therefore, postural control and endurance of ambulation can
be deteriorated with ITB due to impaired gait and balance control. There
may be few cases where spastic muscle hypertonus or muscle spasms interfere with volitional control, thus impeding or reducing gait performance.
Under these conditions it may be worth considering ITB in ambulatory
patients. However, it is strongly recommended to implant the subdermal
pump only after an extended test period with ITB administered by external
pump via temporary lumbar catheter. This is necessary in order to familiarise the patient with the effects that are to be expected and in order to test if
ambulation can be maintained despite ITB doses that are sufficient to control
spasticity. At the same time this may be a first step towards finding of the
individual optimal dose and patient’s expectations can be adjusted prior to
more invasive steps and costly implantation.

Clinical Management of Spasticity and Contractures in Spinal Cord Injury 157

Irrespective of the issue of ambulation, intrathecal test doses should be
applied in all cases where ITB is considered in order to ensure treatment
efficacy prior to implantation. Usually a test dose of 50 ug is applied as an
intrathecal bolus via spinal tap and the patient must be monitored for respiratory depression or arrythmia during the following 6 hours. If no effect is
seen the procedure will be repeated on the following days, increasing the
dose in 25-ug steps up to 75 ug (second day) and if necessary to 100 ug (third
day). This scheme relies on experience from a large multi-center study when
ITB was first introduced [115]. If there is no clear or beneficial effect at a dose
of 100 ug it is unlikely that the treatment with ITB will be successful.
While highly effective, ITB bears a risk of various complications.
Implantation and management of ITB should therefore be confined to dedicated centers with experienced teams. There should be an emergency service available or accessible round the clock, as malfunction of ITB can be
associated with sudden baclofen withdrawal, putting patients at vital risk
if untreated. There are early as well as recent reports indicating a considerable complication rate of 0.011 per month, or 12–13% per year. Of these complications, the majority (78%) were related to catheter malfunction [115,127].
Another more recent and prospective study presented an even higher frequency of adverse events, distinguishing surgical (53%), device-related (29%,
predominantly catheter dysfunctions), and drug-related events (18%) [128].
Drug-related side effects and complications usually comprise drowsiness,
nausea, hypotension, and respiratory depression. They occur mostly during testing of ITB when bolus is applied or when adjustments are made to
the pump settings. There are reports about development of tolerance [129].
Another severe complication may arise in malfunction of the pump or, more
frequently, in dislocation or disconnection of the catheter. In these cases,
withdrawal reactions can lead to malignant muscle hypertonia with hyperthermia, hyperreflexia, potential autonomic instability, seizures, and hallucinations. Patients should be taken under surveillance and treated by oral
administration of the drug.
There is slow development of drug tolerance even with ITB, as was shown
in an early prospective long-term studies [115] and in more recent studies
[115,127] where the authors of this retrospective analysis found an initial
development of tolerance during 5 years but that the mean applied dosage of
baclofen stabilised after 5 years at a dosage of ca. 500 ug/day. No significant
increase in dosage was found thereafter.

For a good overview of ITB management, technical details and complications, side effects, and special considerations the reader is referred to the
recent publication by Khurana [100].
6.3.8 Focal Anti-Spastic Pharmacotherapy: Chemodenervation
An indication for chemodenervation exists in focal spasticity causing harm
or showing progression. Physical management as part of good nursing care,


158

Neurological Rehabilitation

physiotherapy, occupational therapy, exercise, stretching and strengthening
of limbs, splinting, and pain relief are the basis of spasticity management
[78]. Therefore, this should be provided as a basis of any additional treatment
with more invasive techniques, such as chemodenervation. Local spasticity may benefit from oral anti-spastic medication but should preferrably be
treated with focal therapy of the spastic muscle(s), thereby avoiding systemic
side effects. The most common measures are botulinum toxin A (botulinum
neuro toxin, BoNT) and motor point and nerve block by phenol/alcohol as a
cheaper alternative, though potentially at the cost of more local side effects
and pain. Early RCT have proven efficacy of BoNT in the treatment of spasticity in etiologies other than SCI [130,131]. In a RCT comparing BoNT to phenol blockade of the tibial nerve to treat spastic foot after stroke, BoNT was
superior to phenol [132]. This and the other controlled trials together with
the potential local damage that can be inflicted by the injection of alcohol or
phenol has nowadays also led to primary use of BoNT in the treatment of
focal spasticity in SCI. Despite its recognition and acceptance in the management of local spasticity in SCI [78], there is a lack of high-quality evidence
for its efficiency in SCI spasticity, as is revealed by recent systematic review
[133]. Although this literature search, looking at management of spasticity in
a sample population with a majority of SCI patients, identified 9 studies on
BoNT and 10 on phenol/ethanol, none of them were RCT and none of them
were adequately powered. This is true for many studies on the treatment of
spasticity, often due to poor study design, low numbers, a variable management approach, and diverse, non-standardised treatment schemes. As in

other etiologies, improvement of function is difficult to show or achieve [134],
which is, among other reasons, due to the fact that reduction of spastic muscle
hypertonus, as often used as a primary outcome, does not directly translate to
an improvement in function. Furthermore, motor dysfunction in spastic paresis is usually mainly caused by weakness and the other ‘negative’ features of
upper motoneuron syndrome, and not by muscle overactivity [19].
The effect of local treatment with BoNT injected into spastic muscles
causes local weakness via blockade of the neuromuscular junction. The toxin
is internalised by the presynaptic motoneuron, where it inhibits the release
of acetylcholine [135]. The effect of injections is time-limited due to collateral sprouting and regrowth of nerve endings and formation of synapses,
thus the treatment must usually be repeated after 2–6 months. International
guidelines recommend a combination of botulinum toxin injections and
physiotherapy. Phenol and ethanol produce neurolysis when injected close
to the nerve endings that supply spastic muscles. Injection of these agents
causes denaturation, which disrupts neural transmission and subsequently
diminishes muscle activity resulting from central disinhibition. Adverse
events are directly related to the mechanism of action, i.e., muscle weakness
reducing or disturbing functional abilities with both drugs and, additionally, dysesthesia or denervation pain in treatment with ethanol/phenol.


Clinical Management of Spasticity and Contractures in Spinal Cord Injury 159

Specific indications and aims for anti-spastic treatment with focal chemodenervation include: functional improvement (enhanced speed and
mobility, quality, or endurance of gait or wheelchair propulsion, transfers,
dexterity and reaching, and sexual functioning), symptom relief (pain and
muscle spasms, allow wearing of splints, improved hygiene, prevention of
contractures), postural improvement (enhanced body image), decreased care
burden (alleviation of dressing and hygiene processes, positioning for feeding, etc.), enhanced service responses (prevention of need for unnecessary
medication and other treatments, facilitation of therapy, delay or prevention of surgery) [78]. It must be remembered that the treatment of spasticity
is physical primarily in order to influence and control consecutive biomechanical changes. A programme of physical treatment should be established
before and continued during and after pharmacological intervention. Muscle

stretching improves the therapeutic effect of BoNT and vice versa [136], but,
as in other areas of the field, RCT to produce high-level evidence are lacking.
Selection of injection points, dosage, and injection of BoNT should be done
by trained and experienced physicians. Gaining the skills requires time and
commitment. The placing of the injection should be guided by ultrasound or
EMG, which is recorded from the injection needle or by electrical stimulation
of the muscle at the intended target position. The aim of EMG guidance is to
record muscle action potentials at the intended injection site and assess their
interference pattern on muscular activation [78]. Activation can be difficult to
interpret in view of mass synergies in spasticity. Nevertheless, this will help
to detect and distinguish spastic activity from contracture as has mainly
been shown in the upper extremity in children with cerebral palsy [137–139]
but is commonly used in SCI [133]. It can thereby also serve to focus injections to the endplate [140,141]. Local muscle stimulation by electric pulses
that run through the injection needle can be used to localise the motor point
and thereby likely improve efficacy of BoNT treatment, as is indicated by
the growing body of studies elaborating on improved injection techniques
[140,142–144]. In order to get good results, careful thought and planning is
required. BoNT has a good propensity to seek neuromuscular junctions but
placing the toxin as close as possible may best be achieved by electric stimuli
at the lowest intensities to achieve better results. A high-volume dilution and
an endplate-targeted injection are apparently superior to a low volume and
endplate non-targeted injection, when injecting biceps brachii with BoNT in
patients with spastic hemiparesis [140]. It may be assumed that these results
can be extended to SCI but experimental or clinical data are missing. A number of articles are available to guide the treatment of spasticity with BoNT-A
in particular [145,146]. These highlight the principles of treatment and the
need for patients with spasticity to be managed by a multi-disciplinary team.
They present checklists and extend on the right conditions to obtain optimal
outcomes and, importantly, they assist in patient selection and the organisation of services [78].