Neuromodulation Therapies for Spasticity Control: Now and Beyond
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.302464
Source of Support: None, Conflict of Interest: None
Keywords: Baclofen, botulinum toxin, electrical stimulation, magnetic stimulation, neuromodulation, pharmacotherapy, rhizotomy, spasticity, surgery, tizanidine
Spasticity is one manifestation of hypertonia that is both amplitude- and velocity-dependent. This distinguishes it from rigidity, which is constant and independent of passive motion. Spasticity arises from damage to the corticoreticulospinal tracts while rigidity is caused by damage to extrapyramidal pathways. This presentation of velocity-dependent contraction results from the disinhibition of tonic stretch reflexes. This can be explained by denervation of the monosynaptic reflex arc from upper motor neurons (UMNs), which have a net inhibitory effect. The monosynaptic reflex arc consists of a sensory neuron, the cell body of which lies in the dorsal root ganglion, that receives input from the muscle spindle and synapses onto a lower motor neuron (LMN) in the ventral horn of the spinal cord gray matter. The axon of this LMN then exits the spinal cord through the ventral root, into the spinal nerve to innervate the same muscle. Muscle stretch detected by the spindle feeds directly back to induce rapid contraction of the same muscle without communication with the brain, thus termed a spinal reflex. There are many descending inputs from the cortex that carry signals to the LMNs. One major input is that of the corticospinal tract (CST), which is excitatory and elicits voluntary movement. However, the majority of these descending tracts synapse onto inhibitory interneurons, which go on to inhibit the firing of the LMN. Thus, in its normal state, the LMN is in a state of inhibition. Many of these inhibitory signals originate from the cortex and project down through the reticular formation in the brainstem.
The immediate response to the removal of cortical inputs to the LMN is a drastic decrease in the frequency of action potentials transmitted by the LMN. This decreased activity is known as spinal shock and is characterized by the decreased tone in the muscle supplied by the LMN. As the LMN recovers from the initial “shock” of denervation, the virtually unopposed input from the sensory neuron carrying signals from the muscle spindle increases the frequency of action potentials transmitted by the LMN to even greater than before the UMN injury, which leads to more frequent muscle contractions and increased muscle tone., Spasticity is a more frequent sequela of more proximal central nervous system (CNS) damage, i.e., more common with stroke, cerebral palsy (CP), or cervical spinal cord injuries than with thoracic or lumbar spinal cord injuries. Thus, it is also more often observed in tetraplegia than paraplegia. It is also more commonly seen in patients with complete spinal cord injury than partial. Another condition that often causes spasticity is multiple sclerosis (MS); as progressive CNS damage and denervation accumulates, the resulting spasticity increases., Even in nonprogressive conditions such as stroke, anoxic brain injury, traumatic brain injury (TBI), and spinal cord injury, spasticity is often observed to worsen over time.,
Clinically, spasticity can present as anything from a mild increase in muscle tone, barely perceptible on exam, to increases in a tone so severe that joints are almost completely immobile. It can interfere with patients’ ability to carry out many tasks related to daily function and lead to immobility which has its own set of complications related to pressure sores and resulting in an increased risk of infection. Joint dislocations and subluxations as well as heterotopic ossification, most commonly of the hips, elbows, shoulders, and knees are additional painful long-term complications in patients with spasticity. In some situations, however, a patient's spastic rigidity may be relied upon for support in standing or ambulation in the absence of normal voluntary limb mobility.
Spasticity is a clinical diagnosis, which is often supported by the patient's history of a denervating insult such as those listed in the previous section. Removal of UMN inhibitory inputs yields exaggerated spinal reflexes as well as spasticity. Spasticity is a positive sign of UMN damage and is often accompanied by other positive signs such as hyperreflexia and upgoing plantar reflexes (Babinski reflex). There are also negative signs of UMN injury such as voluntary muscle weakness and loss of fine control.
Certain muscle groups exhibit spastic hypertonia more often than others, such as shoulder adductors, forearm pronators, and elbow, wrist, and finger flexors., In the lower extremities, hip adductors, ankle plantar flexors and invertors, and toe flexors and adductors most often exhibit hypertonia in spasticity while either knee flexors or extensors may be affected. This pattern of asymmetric involvement in spasticity leads to characteristic posturing which is easily identifiable on examination such as a clenched fist with the fingers wrapped around the thumb, and equinovarus deformity of the foot.
The characteristic examination finding is a “catch” during passive motion of the affected muscle group due to the unopposed inputs from the spindles sensing muscle stretch. This catch or resistance will increase as the speed of passive movement is increased. Affected muscle groups may also remain in a baseline state of constant contraction while also remaining sensitive to passive stretch. The constant contraction component is referred to as spastic dystonia. The contraction of antagonist muscle groups when the patient attempts voluntary movement may also be observed. These are termed spastic co-contractions.
The most universally accepted scale for grading spasticity is the modified Ashworth scale, which is a five-point numeric scale that ranges from 0–4. A grade of 0 corresponds to a normal tone or no detectable spasticity. Grade 1 is when a “catch” is elicited on examination of the affected muscle group, but the majority of the range of motion (ROM) does not demonstrate increased resistance to passive movement. In grade 2 spasticity, passive motion by the examiner is still relatively easy for the examiner, but there is increased resistance throughout the majority of ROM. Grade 3 is considerable resistance making passive motion difficult, and grade 4 is a rigid joint either in flexion or extension.
Once again it is important to emphasize that all of these diagnostic and grading criteria are determined by physical examination; no diagnostic testing or imaging is required. Quantifiable biomarkers are being investigated, but remain experimental and have not been adopted widely as part of clinical practice. These techniques, especially those that require complex setups to collect patient data, might only prove useful in research settings. Those which are relatively easy to implement and have the potential to direct treatment or monitor progression in a way that will affect the goals of therapy may demonstrate clinical efficacy. Some studies have shown that clinical scores, such as the modified Ashworth scale do not correlate well with more objective measures of the neural component of spasticity such as electrophysiological measurements of muscle stretch reflexes., Measurement of muscle force generated in response to passive stretch has been investigated as a method of quantifying spasticity that would be relatively easy to obtain in the clinical setting., Mechanical models have also been used to represent spasticity as a mechanical impedance to movement about a joint. These measurements correlate closely with clinical assessments and are relatively easy to implement while allowing for much higher resolution than a 5-point scale, possibly allowing clinicians to more accurately follow treatment efficacy or disease progression.
Once the diagnosis of spasticity has been made, decisions must be made about the treatment plan and goals for the patient. These decisions are affected by the etiology of the spasticity, the severity of patient comorbidities, the patient's social support, and the age at which the patient is presenting. As is the case with the majority of medical conditions, the usual course of treatment begins with more conservative therapies (medical management) and then progresses to more invasive treatments as the initial treatment approaches prove inadequate for the patient.
Along with oral pharmacotherapy, the techniques of physiotherapy is the most widely used procedure for spasticity management largely since they can be relatively effective, inexpensive, and noninvasive. It is important to begin therapy as early as possible to prevent the development of permanent contractures and deformities. The initial therapy for the majority of patients consists of stretching, casting, direct tendon pressure, biofeedback, application of heat and cold, electrical stimulation, and vibration. These strategies are often organized under the category of physiotherapy and are the least invasive presenting the least risk to the patient. Physiotherapy works to counteract the muscle overactivity and muscle shortening largely through direct stretching exercises and strengthening antagonist muscle groups.
Other methods of non-pharmacologic conservative therapies include electrical stimulation, transcranial or transspinal magnetic stimulation, and extracorporeal shockwave therapy. Neuromuscular electrical stimulation (NMES) has been used by therapists as a strategy to reduce spasticity, though clinical trials have produced contradictory results. A systematic review of the use of NMES in poststroke patients with spasticity performed in 2015 compiled the results of 29 such clinical trials and examined outcomes related to spasticity, quantified using the modified Ashworth scale and ROM assessed with a goniometer. Of all the trials, 942 patients were included and significant reductions in spasticity, as well as increases in ROM, were identified. An alternative method of transcutaneous electrical stimulation used in the setting of spasticity applies current through acupuncture needles. The addition of a small electrical current is thought by acupuncturists to augment the effects of traditional acupuncture therapy. In the setting of spasticity treatment, it may be functioning similarly to NMES, but instead of the transcutaneous electrical stimulation of NMES, the current is applied through the acupuncture needles piercing the skin. A comprehensive meta-analysis of 22 trials, from both English and Chinese databases, involving 1425 poststroke patients found that there was a statistically significant positive effect of electroacupuncture as adjunctive therapy for patients receiving more traditional physical therapy interventions for lower limb spasticity. Improvements were measured using the modified Ashworth scale for spasticity and the Fugl-Meyer Assessment of Sensorimotor Recovery for overall motor function. The results of the included trials were highly heterogeneous even for lower limb spasticity, and no significant effect was observed in upper extremity spasticity treatment. Improved standardization of electroacupuncture delivery, as well as comparison to transcutaneous NMES, would be beneficial in further investigations of the efficacy of this treatment.
Both transcranial and transspinal therapy are currently being used, but success rates vary widely. One meta-analysis of six clinical trials comprising of 149 patients found that transcranial magnetic stimulation (TMS) for poststoke spasticity lead to no statistically significant improvement versus simulated therapy, while patients with spinal cord injury did realize statistically significant benefit from several modalities of magnetic stimulation including repetitive high-frequency stimulation and intermittent theta-burst stimulation (iTBS) to the primary motor area of the affected leg., iTBS is administered at lower frequencies and shorter durations than traditional TMS and is often better tolerated by patients. A second systematic review of TMS therapy for the treatment of poststroke spasticity included 273 patients also concluded that there is insufficient evidence to support this therapy for improving spasticity poststroke.
Extracorporeal shock wave therapy (ESWT) has recently gained favor as a treatment for many musculoskeletal issues and even wound healing. Its use as an effective treatment for spasticity has been demonstrated by multiple randomized clinical trials with very few adverse effects of ESWT and long-term durability of positive results recorded in the literature at long-term follow-up., Even in light of these findings, it is important to note that the trials of ESWT performed thus far have included relatively small numbers of patients, and the mechanism by which shock waves affect tissue healing or any variables related to spasticity remain unknown. Further, a meta-analysis of 25 studies of the overall effectiveness of physiotherapy for the treatment of spasticity in patients with MS found the strongest evidence supporting exercise therapy and robot gait training as beneficial for patients’ spasticity outcomes such as acute mobility improvement, follow-up measurements, and patient safety. This meta-analysis also included studies of electrical stimulation, and shock wave therapy, but identified less evidence to support the efficacy of these treatments.
In addition to being viable as a treatment for spasticity in itself, physiotherapy is also crucial for rehabilitation following surgical intervention.
Baclofen (β-[chlorophenyl]-γ aminobutyric acid) is an agonist at bicuculine-insensitive receptors in the posterior dorsal horn of the medulla as well as many other locations in the central nervous system, also known as gamma-aminobutyric acid (GABA)-B receptors. Its effect on spinal and supraspinal synapses is overall inhibitory via the restriction of presynaptic terminal calcium influx reducing presynaptic excitatory neurotransmitter release., This central inhibitory action renders it an effective treatment for the disinhibited monosynaptic reflex in spasticity. In multiple double-blind crossover placebo-controlled trials baclofen has been shown to significantly improve spasticity and spasms and it is generally accepted as being one of the most effective oral spasticity medications, however, this finding is not universal, and the measures used to determine spasticity improvement and the duration of treatment vary widely between studies.,,,
Other medications have also been shown to be effective treatments for spasticity, most commonly tizanidine. Studies have shown that baclofen and tizanidine are equally effective, but baclofen remains the most frequently prescribed medication for spasticity.,,,, Both medications have also been shown to not bring about long-term functional benefit for patients, but tizanidine has been reported to cause less weakness than baclofen.,
Gabapentin, clonidine, diazepam, and dantrolene have also been used to treat spasticity with positive results. Dantrolene acts directly on skeletal muscle by reducing the excitation-contraction coupling between actin and myosin. Clinical trials for spasticity have had mixed results and no long-term functional improvement has been demonstrated with dantrolene. Diazepam is just as effective as baclofen and tizanidine in spasticity. Gabapentin has also been found to bring about a statistically significant decrease in spasticity. No controlled studies are available with clonidine for the treatment of spasticity.
Apart from oral therapy, botulinum toxin injection is commonly used as a treatment for spasticity. Botulinum toxin has eight different serotypes: A through H. Only A and B have been used to any significant degree in clinical practice due to their longer biological half-lives. Botulinum toxin A induces muscle paresis from 2–5 days postinjection and muscle function begins to return after 2 to 3 months. The toxin is directly injected into the affected muscles and works to block the release of acetylcholine at the neuromuscular junction to reduce muscle contraction. It has been shown that botulinum toxin injection reduced spasticity relative to placebo in multiple trials of patients after stroke, spinal cord injury, TBI, and in those with CP.,, It is common practice to follow botulinum toxin injection with adjuvant physiotherapy techniques to take advantage of the window of opportunity when the patient's tone is reduced to further combat the development of contractures. Both casting and taping post-botulinum toxin injection have been shown to improve ROM in patients more so than toxin injection alone, and casting was more effective than taping. Evidence suggests that other, less intuitively related therapeutic techniques such as electrical stimulation and extracorporeal shock wave therapy can also significantly improve the effects of botulinum toxin injection on spasticity, ROM, and gait. The combination of botulinum toxin injections with various oral pharmacotherapies for spasticity has also been investigated. Most of these studies have combined botulinum toxin injections with either oral baclofen or tizanidine, and these are the most commonly encountered combination therapies clinically. Though baclofen is still the most commonly used medication in this situation, it has been shown that tizanidine is more effective at reducing spasticity in combination with botulinum toxin, determined by the modified Ashworth scale, and has fewer side effects.,
Surgical management of spasticity falls into three major categories: implantation of a device to deliver a steady dose of medication, physically severing nerves, or implantation of electrical stimulation devices. The etiology and severity of spasticity as well as accompanying disability such as paraplegia and dystonia help to determine if any of these therapies are warranted or will be effective in each patient. For example, some rely on a component of their spasticity for support when ambulating, and while surgical management may more definitively treat their spasticity, this may unmask severe underlying weakness which would render the patient less able to carry out their activities of daily living. As with most conditions, it is usually the case that patients turn to surgical management when physiotherapy and medical management cannot adequately control the disease. While none of these broad categories of surgical intervention are without the potential for adverse effects, the benefits must outweigh the risks.
Implantation of an intrathecal baclofen delivery system is considered when patients do not experience adequate tone reduction on maximal oral therapy with baclofen or other agents, or when the side effects of therapy become unbearable. Up to 25–35% of patients with spasticity of cerebral or spinal origin will fall into these categories. Direct delivery of the medication to the spinal cord allows for fine dose titration to minimize side effects while realizing maximum benefit. Also, unlike relatively permanent, ablative surgical strategies such as selective dorsal rhizotomy, the dose can be reduced to subtherapeutic levels quickly if necessary, functionally returning the patient to a pre-therapeutic state.
Briefly, the process of pump implantation involves the insertion of the drug delivery catheter, placing the pump/drug reservoir in a subfascial pocket, and tunneling from the catheter implantation site to the pump pocket to allow for the connection of the catheter. The catheter is placed by feeding it through a Tuohy needle inserted through the dura, often in an oblique, paramedian trajectory, such that the catheter will lie at the levels deemed appropriate by preoperative testing (usually C5–T2 for spastic tetraplegia or T10–T12 for spastic diplegia). Purse-string sutures are placed around the Tuohy needle to reduce the chances of cerebrospinal fluid (CSF) leak. These are tightened around the catheter and tied once the Tuohy needle and catheter stylet has been removed. The catheter is then secured in place with an anchoring device and suture. The subfascial pump/reservoir pocket is usually made on the patient's anterior left abdomen, and a tunneling device is used to create a tract for the catheter to be passed from the catheter insertion site on the patient's back to the subfascial pocket on the patient's abdomen. Once the catheter has been connected to the pump, CSF return at the pump's catheter access port is confirmed, and the pump is implanted in the pocket and all incisions are closed.
Intrathecal baclofen therapy is highly successful; studies have consistently reported very high success rates in spasticity reduction, often >85%.,, This is in part because, in the majority of studies, implanted intrathecal infusion pumps are generally only provided for patients who are responsive to a test injection of baclofen. Once again; however, as with other treatments, there is not much evidence for long-term functional improvement and increased quality of life., Based on a review of these studies, it seems that this is largely since many patients have multiple other functional limitations which continue to be an issue, both physically and psychologically, even after spasticity is quite effectively treated. Interestingly, it has been reported that after chronic treatment with intrathecal baclofen, there can be some lasting effects on tone reduction even after therapy is discontinued. Dose reduction with the same therapeutic benefit over time has also been reported, though it is more common than some tolerance develops within the first 2 years of treatment, then stabilizes thereafter., Though the monetary cost of pump implantation is significant and transient adverse effects of treatment such as sleepiness and weakness are not uncommon, cost-benefit analyses have consistently supported the use of intrathecal baclofen as net beneficial in comparison to alternative therapies.,
Virtually the only alternative to baclofen for intrathecal delivery is morphine, though its side effect profile including somnolence, vomiting, and confusion, as well as its inferior efficacy, mean it is seldom used. Tizanidine, currently, has no Food and Drug Administration (FDA)-approved intrathecal formulation and very few studies have been performed of intrathecal tizanidine in animal models of spasticity and none thus far in humans.
A selective dorsal rhizotomy is by far the most common of the neuro-ablative procedures for spasticity treatment. Other less common procedures include myelotomy, cordotomy, and peripheral neurectomy which will each be briefly discussed. The effects of these procedures are virtually irreversible, which can be either an advantage or disadvantage depending on the result. All involve lesioning of nerves either by direct transection, thermal, or chemical injury.
The process of selective dorsal rhizotomy involves positioning the patient prone and often in a Trendelenburg position to minimize CSF loss once the spinal dura is opened. Electromyographic (EMG) monitoring of the muscles of the lower limb is also key to enable the surgeon to stimulate and select which nerves to transect unless multilevel laminectomy or laminoplasty and direct visualization of nerve roots are to be performed. This multilevel laminectomy approach has recently fallen out of favor in some centers; now ultrasound localization of the conus medullaris can allow for the procedure to be performed with only a single level laminectomy at the conus itself or caudal to it., Once the laminectomy has been performed and the dura has been opened along its axis, individual nerve roots are isolated and stimulated to confirm their distribution to affected muscle groups. These roots are each divided into several fascicles, some of which are transected based on their ability to elicit sustained discharges in multiple muscles when subjected to direct tetanic stimulation. Preoperative testing using selective botulinum toxin injection can also help to map out the offending nerves. This process is repeated for each nerve root. Once completed, the dura and tissue over the laminectomy defect are closed.
A selective dorsal rhizotomy is more commonly used in patients with spastic diplegia due to CP and has often been shown to be one of the most effective therapies for these patients in reducing tone. Patients have been followed for up to 26 years following selective dorsal rhizotomy with stable tone reduction, a testament to the permanence of this technique. Though the effects on spasticity are often permanent, it is important to note that reduced cutaneous sensation, while common (40–75%), is almost always transient due to redundancies in cutaneous innervation. The permanent nature of the treatment has led some to argue that it is more cost-effective than other surgical methods of spasticity management such as intrathecal baclofen, which requires follow-up procedures such as pump replacements and refills.,
Less commonly performed lesioning procedures for spasticity treatment include myelotomy, cordotomy, and peripheral neurectomy. Myelotomy involves the severing of decussating fibers in the spinal cord by longitudinal incision and is most often used to treat severe pain uncontrolled by less invasive methods. It has shown some utility for durable spasticity reduction but is only suitable for patients with complete paraplegia due to spinal cord injury and no prospect of regaining voluntary motor function below the level of their lesion.,, Cordotomy consists of lesioning the tracts within the spinal cord that carry undesirable signals; once again painful spastic paralysis is where this technique is intended to be beneficial, though it is seldom performed., Peripheral neurectomy is lesioning of the peripheral nerve leading to the spastic muscle. It is less invasive than the other procures listed and poses less risk in that the dura is not opened and the spinal nerves are not accessed, thus it is gaining favor in some circles as a viable treatment for spasticity, especially of the upper limb, where dorsal rhizotomy is of no utility.,,,
Of the three major categories of surgical spasticity treatment, invasive neurostimulation is most infrequently implemented clinically. Some trials of these techniques and devices have been performed, but results have often been inconsistent, and when less technically difficult, less expensive, and less invasive therapies are available which often produce better results, it would follow that such treatment strategies would not be often pursued. The two neurostimulation techniques which have shown some benefit for spasticity treatment in preclinical and clinical trials are spinal cord epidural stimulation and deep cerebellar stimulation.
Early feline trials of epidural spinal cord stimulation for postspinal cord injury spasticity yielded promising results while patient data has often demonstrated unpredictable benefit., Tone reduction has been observed in some trials, especially with stimulators implanted in the lumbar region. Implantation of epidural spinal cord stimulators for spasticity in patients post-spinal cord injury has largely been abandoned as it has demonstrated little benefit for spasticity while being more expensive than competitive therapies. Improvements in technology and continued investigation may allow epidural stimulation to someday enter the standard therapeutic repertoire.,,
It has long been known that lesioning or stimulating the deep cerebellar nuclei can bring about a reduction in tone, while also producing some undesirable effects of dysmetria and tremor. Studies of cerebellar cortical stimulation have demonstrated a similar reduction in tone. Deep cerebellar stimulation attempts to take advantage of this observation as a potential therapy for disorders of increased tone including spasticity. It involves permanent implantation of electrodes in the anterior cerebellum, near the cerebellar peduncles often through a suboccipital or transtentorial approach. One of the largest analyses of patients having undergone deep cerebellar stimulation reported that in 600 patients, there was a significant improvement in spasticity in 66–75% of patients with at least half experiencing at least a 20% improvement. Other functional criteria such as gait were also improved posttreatment. Though still not part of regular clinical practice, once again largely due to relatively effective, less invasive, and less expensive alternatives, deep cerebellar stimulation for spasticity does show promise, especially in patients with CP where upper extremity symptoms and dystonia are significant issues.,
Even with this relatively large number of treatment options available for spasticity, there is still a demand for more effective strategies and the development of novel therapies continues. In addition to the advances discussed herein extant therapies, there are entirely new techniques for the surgical management of spasticity currently being tested. One such technique is peripheral nerve transfer. This involves both hyperselective disconnection of nerves supplying overactive muscle groups and transferring still functional nerves to supply denervated muscles., Following nerve transfer, physical therapy techniques are used to train the patient to use their newly reinnervated muscles. The uneven distribution of spasticity in muscle groups affected by UMN injury makes this strategy potentially more effective as a permanent solution for spastic paralysis with few side effects of systemic treatments and less selective surgical techniques. This technique could theoretically be applied to almost any patient's disease pattern, though the majority of current procedures of this nature have been performed to treat upper extremity spasticity., Clinical trials are currently underway to refine techniques of nerve selection and study the physiological impact of surgical reinnervation.
Spasticity is a common and severe syndrome of disinhibited LMNs. Many options are effective at treating spasticity including physical therapy, oral medications, and neuromodulatory surgeries. Future research is focusing on improving diagnosis and biomarkers for spasticity and treating spasticity while preventing common side effects such as weakness, decreased sensation, and loss of ambulation.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.