Pediatric Annals

The Management of Cerebral Palsy: How Can Neurosurgery Help?

Rick Abbott, MD

Abstract

Cerebral palsy is a disorder of motor control resulting in abnormalities of posture, muscle tone, and motor coordination. It results from mishaps to the motor areas of the developing brain. Cerebral palsy is a clinical syndrome and not a distinct entity. The incidence in infants with birth weights less than 1500 g is almost 30 times greater than among full-term infants.p 1 Etiologies include vascular, traumatic, toxic, metabolic, and perhaps even teratogenic insults to the developing brain. In most cases, the cause is unknown. Children with cerebral palsy may also have other associated handicaps, including visual, hearing, and cognitive deficits.

Cerebral palsy is common and affects 1.5 to 2.5 per thousand live births in the United States yearly. It is estimated that more than 100 000 Americans under age 18 years have some neurologic disability attributed to cerebral palsy. Approximately 25% of these patients are unable to walk and almost 30% are mentally retarded.p 2 The economic impact in the United States in 1992 was $503 000 per new case and the cost to society was approximately $5 billion in 1992 dollars.p 3 The 30-year survival of patients with cerebral palsy is approximately 87%.p 4

Although the brain lesion is static, the resulting movement disorder may change and either worsen or actually get better.p 5 The Collaborative Neonatal Project found that almost 66% of children diagnosed with spastic diplegia and 50% of those diagnosed with cerebral palsy at their first year of life had outgrown motor signs of cerebral palsy by age 7 years.p 6

DEFINITIONS

Cerebral palsy is classified as spastic, atheoid, or a combination of these two. Spastic cerebral palsy is more common, accounting for 65% to 75% of all cases. Spasticity is defined as a motor disorder characterized by a "velocity-dependent increase in tonic stretch reflexes (muscle tone) with exaggerated tendon jerks resulting from hyperexcitability of the stretch reflex as one component of the upper motor neuron syndrome."p 7 In clinical terms, this represents a velocity-dependent increased resistance to passive muscle stretch. Spasticity usually affects lower extremities more than upper. It is characterized by increased stretch reflexes, increased muscle tone, persistent primitive reflexes, and lack or delay of normal motor skills. The muscles are in continuous contraction, which ultimately leads to joint and bone deformities and contractures.

Children with atheoid cerebral palsy exhibit abnormal, involuntary (choreoid) movements. These are due to disordered muscle tone involving the extremities, trunk, and facial muscles. Ataxic cerebral palsy is the least common type. It is caused by disruption in cerebellar pathways and characterized by a lack of coordination. Children with this form of cerebral palsy cannot perform rapid repetitive movements and have a wide-based gait.

Cerebral palsy is also classified according to the extremities: monoplegic, with one extremity; paraplegic, with only the lower extremities involved; diplegie, with primary involvement of the lower extremities and minor involvement of the upper extremities; hEmiplegic, where one side of the body is involved; and quadriplegic, where the entire body is affected. Children born before 32 weeks' gestation with a static encephalopathy will typically manifest spastic diplegia, whereas those born at term will show a spastic quadriplegic pattern.^sup 8-10^

During the latter half of this century, the treatment of spasticity in children has become complex as numerous tools have been added to the physician's armamentarium. Previously, spasticity was treated with range of motion exercises and by surgical repairs of contractures. Pediatricians who care for children with cerebral palsy should be familiar with the current therapeutic options and their side effects.

THERAPEUTIC OPTIONS

Early Surgical Treatment of Childhood Spasticity

Late in the 19th century, surgeons began to…

Cerebral palsy is a disorder of motor control resulting in abnormalities of posture, muscle tone, and motor coordination. It results from mishaps to the motor areas of the developing brain. Cerebral palsy is a clinical syndrome and not a distinct entity. The incidence in infants with birth weights less than 1500 g is almost 30 times greater than among full-term infants.p 1 Etiologies include vascular, traumatic, toxic, metabolic, and perhaps even teratogenic insults to the developing brain. In most cases, the cause is unknown. Children with cerebral palsy may also have other associated handicaps, including visual, hearing, and cognitive deficits.

Cerebral palsy is common and affects 1.5 to 2.5 per thousand live births in the United States yearly. It is estimated that more than 100 000 Americans under age 18 years have some neurologic disability attributed to cerebral palsy. Approximately 25% of these patients are unable to walk and almost 30% are mentally retarded.p 2 The economic impact in the United States in 1992 was $503 000 per new case and the cost to society was approximately $5 billion in 1992 dollars.p 3 The 30-year survival of patients with cerebral palsy is approximately 87%.p 4

Although the brain lesion is static, the resulting movement disorder may change and either worsen or actually get better.p 5 The Collaborative Neonatal Project found that almost 66% of children diagnosed with spastic diplegia and 50% of those diagnosed with cerebral palsy at their first year of life had outgrown motor signs of cerebral palsy by age 7 years.p 6

DEFINITIONS

Cerebral palsy is classified as spastic, atheoid, or a combination of these two. Spastic cerebral palsy is more common, accounting for 65% to 75% of all cases. Spasticity is defined as a motor disorder characterized by a "velocity-dependent increase in tonic stretch reflexes (muscle tone) with exaggerated tendon jerks resulting from hyperexcitability of the stretch reflex as one component of the upper motor neuron syndrome."p 7 In clinical terms, this represents a velocity-dependent increased resistance to passive muscle stretch. Spasticity usually affects lower extremities more than upper. It is characterized by increased stretch reflexes, increased muscle tone, persistent primitive reflexes, and lack or delay of normal motor skills. The muscles are in continuous contraction, which ultimately leads to joint and bone deformities and contractures.

Children with atheoid cerebral palsy exhibit abnormal, involuntary (choreoid) movements. These are due to disordered muscle tone involving the extremities, trunk, and facial muscles. Ataxic cerebral palsy is the least common type. It is caused by disruption in cerebellar pathways and characterized by a lack of coordination. Children with this form of cerebral palsy cannot perform rapid repetitive movements and have a wide-based gait.

Cerebral palsy is also classified according to the extremities: monoplegic, with one extremity; paraplegic, with only the lower extremities involved; diplegie, with primary involvement of the lower extremities and minor involvement of the upper extremities; hEmiplegic, where one side of the body is involved; and quadriplegic, where the entire body is affected. Children born before 32 weeks' gestation with a static encephalopathy will typically manifest spastic diplegia, whereas those born at term will show a spastic quadriplegic pattern.^sup 8-10^

During the latter half of this century, the treatment of spasticity in children has become complex as numerous tools have been added to the physician's armamentarium. Previously, spasticity was treated with range of motion exercises and by surgical repairs of contractures. Pediatricians who care for children with cerebral palsy should be familiar with the current therapeutic options and their side effects.

THERAPEUTIC OPTIONS

Early Surgical Treatment of Childhood Spasticity

Late in the 19th century, surgeons began to report a series of surgically treated patients with spasticity. The largest was Foerster's group of over 150 patients, including 96 children with spastic congenital (or infantile) "paraplegia." The side effects of a complete rhizotomy (cutting of a nerve root within the spinal canal; loss of protective pain sensation and sensory ataxia) and the poor functional improvement (caused by the lack of supportive physical therapy) kept the procedure from gaining popularity despite impressive effects on limb tone.p 11

In the late 1960s, the procedure began to be reexplored in Europe. Knowing complete rhizotomies to be unacceptable, experiments with limiting the degree of sensory root were performed.^sup 12,13^ Gros and his associates demonstrated a decrease in side effects and maintenance of efficacy of sensory rhizotomy when partial lesioning of the roots was performed. This trend in attempting to define the critical mass of sensory fibers, what must be cut to treat the spasticity and what must be preserved to avoid side effects, has continued through the past several decades.

Myelotomy (severing of fiber tracts within the spinal cord) was first described by Bischof in 1951. His initial technique was to perform a sectioning of the cord from its lateral surface centrally, seeking to cut into the intermediate gray matter of the spinal cord.p 14 In 1960, Pourpre advanced the technique by confining his lesion to the spinal gray matter.p 15 Although many variations on this theme have followed, myelotomy is not used widely in the treatment of childhood spasticity, despite some authors describing outcomes that rival other neurosurgical treatments.^sup 16-19^ This is because of its inherent risk of injury to the corticospinal and micturition pathways.

Current Deafferentation Procedures Used for Childhood Spasticity

In 1974, Sindou documented the location of the la sensory fibers (the supposed mediators of the hyperactive reflex arcs responsible for spasticity). These are present at the root entry zone within the dorsal horn for humans.p 20 He then proposed a surgical technique that transected these fibers as they entered the cord. This technique has come to be known as the drezotomy. Lesioning was directed according to a preoperative map describing which non-antigravity muscles were involved by the spasticity. Intraoperatively, sensory rootlets (subdivisions of the sensory roots that normally occur near their point of entry into the cord) were stimulated and those causing contraction of targeted (spastic, non-antigravity muscles) muscles were cut. All of these techniques only sought to differentiate nerve fibers from involved muscles and confine the lesion to the former. None sought to identify sensory fibers that actually mediated abnormal reflexive responses and hypertonia.

Fasano then found that some sensory fibers, when electrically stimulated with 30 to 50 Hz trains, would trigger a one-for-one response in the motor nerve with an associated muscle contraction.p 21 Other sensory fibers stimulated with an identical train would not: instead, he documented a rapid abolishment of the motor response within milliseconds of the initiation of the stimulus train. Fasano went on to describe the former pattern as identifying a circuit that had lost descending inhibitory control and thus was mediating spasticity (an abnormal root/rootlet). The latter pattern identifies normal roots/rootlets. Other patterns of response that he described as abnormal roots/rootlets were the tetanic contraction of muscles lasting longer than the duration of stimulation of the sensory nerve and spread of muscle activation to outside the myotome of the spinal segment whose sensory root was being stimulated, even to another limb. Fasano used these abnormal response patterns to identify rootlets that were candidates to be cut. He coined the term "function dorsal rhizotomy" to describe the technique. This is the rhizotomy technique that most centers in North America used in the 1980s and 1990s as a starting point to treat spasticity.

Physiologic Basis of Current Deafferentation Techniques

It is felt that current deafferentation techniques alter the modulating milieu of the interneuronal pool controlling the reactivity of the alpha motor neurons. The interneuronal pool's primary function is to modulate the pattern and reactivity of the spinal cord's reflex circuitry.^sup 22-24^ This pool is inhibited by sensory nerves coming from the limbs (eg, muscle stretch receptors, joint receptor afferents, and cutaneous receptor afferents) and stimulated by descending fiber tracts from the brain as well as from other sites.^sup 22,23^ Thus, the system is wired so the brain can inhibit the spinal cord's normal, reflexive response to incoming sensory impulses and thereby effect muscle tone. With either a spinal cord injury or brain injury, the loss of descending cord fibers causes an unbalancing of these modulating influences on the interneuronal pool. Because of a relative decrease in the controlling influence for the alpha motor neuron pool sensory input leads to a net amplification of reflexive output. The underlying rationale of deafferentation is that by cutting la sensory fibers, the lack of descending cord fiber input is offset and the alpha motor neurons are brought back into balance.

Selective Dorsal Rhizotomy

Most North American centers performing selective dorsal rhizotomies to treat childhood spasticity have modeled their surgical technique on Fasano's.p 21 Early experience demonstrated that the urologic system's function was vulnerable to operative injury. Because of this, it was advocated to move the site of the operation to the lumbar canal, where the sensory roots could be identified using surrounding bony anatomy. This allowed the identification and preservation of the mid-sacral roots participating in urologie function. Use of multichannel electromyelographic (EMG) recording in monitoring for diffusion of muscle activity during nerve root stimulation has led to decreased reports of postoperative urologie dysfunction.

As centers gained experience with the procedure, modifications evolved. Most report they are cutting fewer rootlets. Typically, between 20% and 50% of the rootlets tested will be cut. This contrasts with up to 80% lesioning density cited in earlier procedures. There are several explanations; anesthetic technique has evolved and, as a consequence, the reactivity of the nervous system being tested has changed. Selection criteria used to label roots/rootlets as abnormal have also changed.

Many centers performing selective posterior rhizotomy on patients with spastic cerebral palsy have now published outcomes. All describe a significant impact on spasticity after rhizotomy and none significant return of spasticity over time. Peter, in discussing 104 children 2 to 12 years of age who underwent selective posterior rhizotomy, found that 95% of his patients had a long-term reduction in tone and this persisted.25 Cahan reviewed the results of instrumented gait analysis in 14 children and found that the EMG signature of spasticity seen preoperatively had disappeared in postoperative testing 6 to 14 months after their surgery.p 26 Park describes rhizotomy as having "always reduced spasticity" in his patients.p 27 Cohen found "an immediate and significant reduction in muscle tone in the lower extremities of every patient."p 28 Steinbok found a significant decrease in tone in the lower extremities of 50 children as measured using a myotome. This decrease persisted for at least the first year after their surgeries. Peacock reviewed 25 patients who had rhizotomies for spasticity using a modified Ash worth scale (Table). He found that all demonstrated normalization of tone with an associated loss of deep tendon reflexes and clonus at an average 8.9 months after their surgeries.p 29 Our 1-year follow-up modified Ash worth scale data found a statistically significant reduction in tone for all 49 patients for every muscle tested. This reduction persisted at their 6-month to 1-year follow-up examinations.p 30 We have since reexamined these patients 5 years after rhizotomy. The tone changes seen at 1 year have persisted. When we have seen return of spasticity, this has always been associated with some noxious stimulus such as a case of otitis or a hip dislocation. Once the painful stimulus had been dealt with, the tone reverted to its post-rhizotomy, premorbid state.

A principle goal for any treatment seeking to normalize muscle tone in a spastic child is to improve the available range of motion in the leg's musculature. Most centers have found that knee extension has improved in patients who have undergone rhizotomy. Improvements in the following groups of muscles have been seen: hamstrings, dorsiflexion of the ankles, the available range of the hip adductors (hip abduction), and plantar flexors. We have found a statistically significant decrease in range limitation in the hip adductors and hamstrings at 1 year post-rhizotomy and this decrease persisted at 5-year follow up.p 30 It should be noted that hamstring muscles appear vulnerable to loss in range over time after initial improvement. This appears to be the case for both diplegic and quadriplegic groups. Although this finding may not be clinically significant in a child who is wheelchair bound, it can have a great impact on ambulation of the higher functioning child. We have found this particularly common during growth spurts and it is problematic in children who are not being closely monitored by trained personnel (ie, therapists or parents educated to a paraprofessional level).

Table

TABLEAsh worth Scale of Muscle Tone

TABLE

Ash worth Scale of Muscle Tone

More important than changes in muscle tone or available range is an understanding of the functional changes that can be expected after treatment of spasticity. Recently, centers have begun to publish short-and long-term functional outcomes of children who have had selective posterior rhizotomies to treat spastic cerebral palsy.p 31 Peacock reported on the ability of nine of his patients to assume and maintain static, functionally important positions such as sitting (five required less support), moving from a kneeling to a two-kneeling position (six improved in this), and rising from a seated to a standing position (four improved).p 29 Berman found that 27 of 29 South African children experienced improvement in their ability to long-sit (sitting square on floor with legs extended straight in front), with nine improving 2 or more grades on a 5-grade scale.p 32 Nineteen of 29 improved in their side-sit (sitting on right or left side of hips with legs folded to side) capability, 24 in their half-kneeling (sharpshooter position) ability, and 15 in their standing ability. We have had the opportunity to evaluate 41 patients post-rhizotomy over 5 years. Of the 11 walkers preoperatively, 2 improved in longsitting, 1 improved and 3 declined in side-sitting, 5 improved in half-kneeling, and 2 improved in standing. With regard to the 11 who preoperatively crawled in a quadruped fashion (crawlers), 7 improved and 1 declined in long-sitting, 9 improved in side-sitting, 10 in half-kneeling, and 9 in standing. Of the 19 not functional in locomotion except, at best, as commando crawlers (non-locomotors), 4 improved in long-sitting with 2 declining, 5 in sidesitting with 3 declining, 8 in half-kneeling with 3 declining, and 4 in standing with 2 declining.

Another means of quantitating functional outcome is computed gait analysis. The major postoperative improvement is an increase in the stride length. This is not surprising given improvements seen in knee extension after rhizotomy and an increase in the amount of pelvic tilt, presumably due to an asymmetrical effect of the rhizotomy with the hip extensors experiencing a greater reduction in tone than the hip flexors.p 33 Two groups were also able to quantitate improvement in ankle dorsi flexion. Adams also found a significant increase in valgus deformity (foot rolled laterally so standing was on its lateral aspect) at terminal stance. Consequently, ankle-foot orthoses were recommended to prevent hindfoot and forefoot eversion. He also recommended the inclusion of a dorsiflexion stop to deal with a weakness in the plantar flexors because this bracing tends to inhibit excessive forward momentum of the tibia during the stance phase.

Although there are many reports on the effects of the surgery on muscle tone, flexibility, gait patterns, and functional positioning, families rarely seek assistance in these areas. Rather, they desire treatment that will allow their child to deal with his or her environment. These sorts of improvements have been graded using the Pediatric Evaluation Disability Index (PEDI) or Functional Inventory Measure for Children (WeeFIM). Dudgeon found that all 16 spastic diplegics had improved PEDI mobility scores and 13 of 16 improved in self-care scores a year after their surgeries.p 34 He also reported that five spastic quadriplegics did not experience impressive gains. Bloom found a significant improvement in PEDI self-care scores for 8 spastic diplegics and 8 spastic quadriplegics.p 35 Nishida, using the WeeFIM, found that spastic quadriplegics improved most in bowel and bladder function, whereas the spastic diplegics showed the greatest gains in mobility index.p 31 Authors have also repotted improvement in upper limb function after selective rhizotomies of sensory roots of the lower extremities. Steinbok found that 26 of 39 patients experienced improvement in upper limb function, and Kinghorn found that, of 7 children with increased tone in their upper extremities of a disabling nature, 6 had improvement in tone to such a degree as to allow improved function.^sup 35,36^

The most common intraoperative complications seen are respiratory.p 30 Any child bom prematurely, especially one with a history of bronchopulmonary dysplasia, is at a fivefold increased risk for developing reactive airway disease. A history of gastroesophageal reflux or aspiration pneumonia is associated with a threefold increased risk of reactive airways. In reviewing our first 60 patients, we found a 22% incidence of problematic perioperative bronchospasm, and it occurred in children with histories or risk factors for reactive airway disease.p 30 Children with more advanced spastic cerebral palsy have an increased risk for aspiration pneumonia.^sup 30-37^ These are typically children with a history of perinatal gastroesophageal reflux or aspiration pneumonia. We have found a 12% incidence of aspiration pneumonia intraoperatively in our first 60 patients and thus begin administering bronchodilators and oral steroids 48 hours prior to surgery.

Hypotonia or "weakness" is frequently mentioned as a complication of selective rhizotomy with a cited incidence of 4% to 100%.p 38 Most larger series report a 2% to 5% incidence of sub-dermatomal hypesthesia after selective dorsal rhizotomy.^sup 39,40^ Dysesthesias (a pins and needles sensation covering the lower legs) are also reported in between 10% and 50%, but most patients report that this resolves within several weeks of the surgery.p 40

Children with cerebral palsy are already at risk for urologic dysfunction. McNeal has reported that up to 36% exhibit symptoms suggesting bladder dysfunction (enuresis, stress incontinence, and dribbling).p 41 Consequently, these children should be screened preoperatively for symptoms of urologie dysfunction and undergo formal testing if symptoms suggest any dysfunction. Also, urologie function should be protected as the nervous system is being operated on. Postoperative urinary dysfunction has been attributed to cutting more than 50% of each of the S2 roots. Currently, we cite a 2% to 3% risk of transient dysfunction: this requires intermittent catheterization for several days to weeks, presumably due to a postoperative cystitis.

Selective dorsal rhizotomy requires spinal manipulation and this, in addition to the known incidence of scoliosis in children with cerebral palsy, puts children undergoing this operation at risk for spinal deformity. Peter has published the most definitive study on this problem, reviewing radiographs of 55 children 1 to 7 years after their operation.p 42 They found 9 with scoliosis, 4 with lordotic curvature, and 5 with spondylolithesis.

Even though spasticity is markedly decreased by the procedure, there is potential for progressive loss of joint mobility postoperatively due to muscle contracture. This can require a tendon lengthening or transfer. Several have reported that 8% to 30% of patients will require secondary tendon operations post-rhizotomy.^sup 43,44^

Hip dislocation has also been seen after posterior rhizotomy. These can be progressive or residual from before the rhizotomy. Other complications include cerebrospinal fluid (CSF) wound leakage, subdural hematomas, and headache.p 28

There have now been numerous outcome studies describing dorsal rhizotomies in children. Although many have called into question the mechanism of action of the surgery, it seems clear that spasticity reliably diminishes. As with many other treatments for these children, attention must now be directed toward determining when rhizotomy is the optimal means of treatment and how electrophysiologic monitoring should best be used.

Intrathecal Baclofen

Parental baclofen (Lioresalp 7) was first introduced in 1967 for the treatment of spasticity of spinal origin, and it rapidly became the drug of choice for this condition. Baclofen (4-chlorophenyl GABA) is an agonist of gamma-aminobutyric acid (GABA), which preferentially binds at GABA B receptor sites. These receptor sites are primarily within the spinal cord's dorsal horn in laminae II and III. GABA B receptors are involved in presynaptic inhibition of sensory afferents. The binding of baclofen at these sites results in a presynaptic inhibition of IA sensory afferents with a resultant decrease in reflexive motoneuron activation. This, in turn, lessens the respective limb's hypertonia. Although there is little doubt as to the drug's effectiveness in treating spasticity, its utility has been limited because it does not readily cross the blood-brain barrier when taken orally. Not uncommonly, when the dose is increased to establish a better response, intolerable side effects such as drowsiness, confusion, and hallucinations result. Because of this, Penn and Kroin investigated using the drug intrathecally. They documented a dramatic drop in limb tone when the drug was administered into the CSF.p 45 They showed that after the spinal cord is injured and descending fiber tracts are lost, GABA B binding sites within the dorsal horn of the spinal cord proliferate.p 46 These studies confirmed that these binding sites lay close to the dorsal surface of the cord and that the drug, once in the intrathecal space, could penetrate in high concentration to these receptors. This results in a dramatic lessening in the amount of drug needed to obtain the desired clinical response.

Albright believes four groups of patients are candidates for intrathecal baclofen.p 47 The first consists of children whose spasticity is functionally disabling, who ambulate, and whose leg strength is poor. Leg weakness is to such a degree as to raise concern that they could not maintain function if spasticity were completely eliminated by rhizotomy. The second are patients older than 16 years who are ambulatory. Here, again, the concern is that the patient's leg strength could be inadequate to maintain function were the spasticity completely eliminated. Older individuals have extreme difficulty in muscle strengthening. Albright's third group of candidates are nonambulatory spastic quadriparetics whose spasticity interferes with daily living and comfort, and the final group are nonfunctional spastic quariparetics who are completely dependent on others for daily care. The pump is inserted in this last group to ease the chores of the patient's caregiver(s). There is some argument over the last two groups as our experience has been that many with such conditions respond nicely to selective dorsal rhizotomy. We would treat children in these last two groups with rhizotomy if they had pure spasticity and would use intrathecal baclofen if they had mixed cerebral palsy manifesting both spasticity and rigidity.

When a child's tone examination has identified them as a potential candidate, they are brought to the hospital for a trial injection of intrathecal baclofen. After completing a tone examination of the arms and legs, a 50 g bolus is injected into the lumbar thecal sac. A tone examination is then performed every 2 hours over the next 8 hours. A positive response is defined by a significant drop ( 1 or more points on the Ash worth scale) in the average muscle tone of the legs that is sustained over two or more examinations. If this response fails to occur, the injection dose may be increased to 75 g the next day, and if this fails, a final trail injection of 100 g may be given. Failure may not necessarily mean that the patient is unresponsive to the drug. A patient will not respond to an injection into the subdural space and alteration in muscle tone will not be appreciated in a patient with extreme muscle contracture, which precludes limb movement and hence tone examination.47

Once the efficacy of treating the child's hypertonia with intrathecal baclofen has been established, he or she is taken to surgery for implantation of the pump. The pump comes in two sizes, a 10- and a 18-cc capacity, each with the diameter of a hockey puck. Because of their sue, there is a body weight minimum that generally excludes children under 3 years. The pump is implanted in the anterior abdominal wall, lying on the anterior rectus sheath. Connected to it is a silastic catheter that runs subcutaneously to the spine and whose outlet rests over the conus of the spinal cord within the CSF space. The pump's parameters are set and altered using a computer with a broadcasting antenna. The initial daily dose is similar to the dose of the successful trail injection. This dose can be increased daily by 20% to 30% until the desired effect is obtained. The child is kept in the hospital several days while the dose is being adjusted and the child is recovering from surgery. The pump's reservoir will typically require refilling every 1 to 3 months and the pump's battery is expected to last from 4 to 5 years, after which the pump must be replaced.

Although the bulk of outcome studies for intrathecal baclofen have been performed on adults, there are now several reports of outcomes in children. Twenty of Muller's 72 patients were children and this paper reports that 90% to 95% of these 72 patients experienced improvement.48 Albright found that 37 children had a significant decrease in muscle tone in both arms and legs.49 He has also documented similar functional improvement as would be expected in children undergoing selective dorsal rhizotomy. Campbell has also documented functional improvement in patients receiving intrathecal baclofen but also stated that only 10% to 30% felt that this improved their quality of life.50 Recently, a multicenter drug trial has been completed, and the FDA has approved the drug's use intrathecally in children to treat spasticity of cerebral origin. The results of this study group will soon be published. The two principle complications seen with this treatment are iatrogenic drug overdose (typically due to excessive drug delivery after altering delivery rate or changing drug concentration) and infection. The drug rate is approximately 5% and generally consists of an infected seroma around the pump. If meningitis occurs, it generally happens during drug trials when an externalized catheter is being used to deliver the drug.

Patients receiving too high a dose will generally act lethargic or somnolent, and their extremities (especially their legs) will feel floppy due to hypotonia. This reverses within hours of decreasing the delivery rate. If the delivery rate is not decreased in this situation, the child can go on to become unresponsive as the hypotonia progresses and ascends. If the hypotonia is allowed to progress, respiratory collapse will ensue and ventilation will need to be supported with an endotracheal tube and mechanical ventilation. After supporting ventilation, the pump is shut off, its reservoir emptied, and 30 to 40 cc of CSF is removed. Some of this fluid should be sent to the laboratory to confirm an elevated baclofen concentration.

This therapy will probably become much more common as its efficacy in treating children with mixed cerebral palsy becomes established. Additionally, given die ease of establishing candidacy and the simplicity of implantation, many will be attracted to its use for treating children with spasticity. However, this is a tool, one of many, to be used in the appropriate setting. It is best used in a multidisciplinary setting in which many treatments for children with cerebral palsy are available and each is given thoughtful consideration.

Peripheral Neurotomy

Spasticity in a muscle can be attacked peripherally by reducing the muscle's nerve supply. The rationale of such treatment is to weaken a spastic muscle by diminishing its innervation. This can be accomplished by surgically exposing the nerve so it can be partially or completely cut, percutaneously injecting ethanol or phenol around the nerve to partially sclerose it, or by blocking the muscle's neuromuscular junctions with botulinum toxin injections.

Open neurotomy or neurectomy requires surgical exposure of the nerve at the point where it gives off branches to the targeted muscle.51 Because the nerve is directly visualized, the surgeon can control exactly what gets cut, thus minimizing collateral injury to nontargeted tissue. Once exposed, the branches are stimulated electrically, allowing the surgeon to confirm what muscles are innervated by the nerve branch and to establish the force of muscle contraction. The patient's preoperative examination has been used to formulate a surgical goal for weakening the muscle, and this guides the surgeon in how much of the nerve will be cut. The lesioning proceeds in a stepwise fashion with frequent restimulation of the nerve to allow tracking the effect of lesioning on strength of contraction.

Percutaneous neurotomy is performed using an injectable, sclerosing agent that penetrates the nerve sheath to partially destroy the nerve supply to a spastic muscle. A needle is inserted into a region known to contain the targeted nerve. The location is then optimized using electrical cunent through the needle to confirm contact with the targeted nerve. The sclerosing agent (either ethanol or phenol) is then injected. The amount that can be injected at any one setting is limited by the systemic side effects of these agents. The injection may take place at several points within the targeted muscle (motor point blocks) in an attempt to decrease the number of gamma fusimotor fibers (these fibers are affected preferentially by the agents) or more proximal (chemical neurolysis) for a more diffuse effect on the nerve's pattern of innervation. Smaller fiber types within the nerve seem to be more vulnerable to these agents. With time, sprouting of remaining nerve fibers within the muscle results in a variable return of the hypertonia of the muscle. Although there are many reports on the use of phenol as the sclerosing agent, it can potentially be much more injurious than ethanol to surrounding, non-targeted tissue. Additionally, it poses a great risk to the patient should it be injected into a blood vessel. For that reason, ethanol is favored by some for nerve blocks in children.

Botulinium Toxin

More recently, botulinium toxin (BOTOX) has gained popularity over other forms of percutaneous neurectomies.51 Although this is an "off-label" use of the agent, several reports of its successful use for childhood spasticity are now available. It ineversibly binds at the neuromuscular junction to prevent release of the neurotransmitter acetylcholine. This results in denervation of the muscle, with a disuse atrophy then occurring. In general, only transient benefits, lasting up to several months, can be expected after an injection. This is similar to derivatives for other forms of percutaneous neurectomies.52,53 Neurons send new axons into the denervated region and new neuromuscular junctions are formed, resulting in a reinnervation of the muscle. Consequently, BOTOX, as well as other types of percutaneous neurectomies, are typically used either to evaluate how denervating a spastic muscle would affect the limb's function or to temporarily counter the effects of a muscle's hypertonicity on developing limb function. Because some form antibodies to the toxin, the maximum dose of BOTOX currently recommended by its manufacturer is 6 U/kg every 6 months. Because the drug is frequently used as a temporizing agent if there is little concern about the child developing antibodies (and thus resistance to treatment), some are using higher doses on the order of 8 U/kg for 6 months. If all targeted muscle cannot be treated by a safe dose, ethanol can be used to neu - roiyse the nerve supply to untreated muscles.

Neurectomy

The obturator nerve is the most frequent to undergo open neurectomy. This is done to halt progressive subluxation of a hip from hip adductor spasticity. In the past, complete neurectomies with adductor myotomies were done to treat hip adduction deformity. It was soon found that this resulted in persistent hip abduction contracture due to loss of opposing movement.53 Consequently, it was replaced with a partial neurectomy of the anterior branch of the obturator nerve. Most (80% to 86%) with these partial obturator neurectomies have improvement or resolution of hip adduction deformity after partial obturator neurectomy.54,55

Several European groups have series of patients who received neurectomies to treat other spastic muscles.56 Spasticity of the triceps surae and toe flexor muscles can be treated using selective lesioning of motor branches of the tibial nerve. Selective neurotomies are also useful for spastic muscles of the upper extremities.

CONCLUSION

Neurosurgery has a role in assisting in the management of children with cerebral palsy. It can offer many treatments; however, not all are appropriate in all settings. The philosophy of our clinic is to arrive at that treatment or treatments that will be least invasive, provide the greatest functional benefit, and create the least dependency on us. We need to work closely with the pediatricians who care for children with lifelong disabilities.

REFERENCES

1. Albright AL Spasticity and movement disorders in cerebral palsy. J Child Neurol. 1996;11:S1-S3.

2. Kuban KCK, Levitón A. Cerebral palsy. N Engl] Med. 1994;330:188-194.

3. Economic coses of birth defects and cerebral palsy - United States, 1992. MMWR. 1995;44:695-699.

4. Cirtcton JU, Mackinnon M, White CP. The life expectancy of persons with cerebral palsy. Dev Med Chad Neurol. 1995;37:567-576.

5. Baratas G, Taft LT. The early signs and differentiated diagnosis of cerebral palsy. Pediatr Mn. 1986;15:203-214.

6. Nelson KB, Ellenberg JM. Children who outgrew cerebral palsy. Pediatrics. 1982;69:527-536.

7. Lance JW. Symposium synopsis. In: Feldman RG, Young RR, Koella WP, eds. Spasticity: Disordered Motor Control. London: Yearbook Medical Publications; 1980:485-500.

8. Stanley F. Perinatal risk factors in the cerebral palsies. In: Stanley F1 Alderman E, eds. The Epidemiology of Cerebral Paktes. Philadelphia. Pa: Lippincott; 1984:98-1 1 5.

9. Stanley F. Prenatal risk factors in the cerebral palsies. In: Stanley F, Alderman E, eds. The Epidemiology o/ the Cerebral Palsies. Philadelphia, Pa: Lippincott; 1984:87-97.

10. Stanley F. Postnatal risk factors in the cerebral palsies. In: Stanley F, Alderman E, eds. The Epidemiology of the Cerebral Palsies. Philadelphia, Pa: Lippincott; 1984:135-149.

11. Foerster O. On the indications and results of the excision of posterior spinal nerve roots in men. Surg Gynecol Qfeset. 1913;16:463-474.

12. Gros C, Ouaknine G, Vlahovitch B, Frerebeau P. La radicotomie selective postérieure dam le traitement neuro-chirurgical de !'hypertonic pyramidale. Meurochirurgie. 1967;13:505-518.

13. Ouaknine GE. (Surgical treatment of spasticity - role of selective posterior rhyzotomyl. Union Med Can. 1980;109:1424-1444.

14. Bischof W. Die longitudinale Myelotomie. ZH Neurochir. 1951;1 1:79-88.

15. Pourpre MH. Traitement neuro-chirurgical des contractures chez les paraplégiques. Neurochirurgie. 1960;6:229-236.

16. Bischof W. Zur dorsalen longitudinalen myelotomie. Zentralb! Neurochir. 1967;28:123-126.

17. Laitinen L, Singounas E. Longitudinal myelotomy in the treatment of spasticity of the legs. J Neurosurg. 1971;35:536-540.

18. Laitinen LV. Longitudinal myelotomy for spasticity. Iru Sindou M, Abbott R, Keravel Y, eds. Neurosurgery for Spasticity. New York, NY: Springer-Verlag; 1991:183-186.

19. Yamada H, Miyahara K, Karneya T, Oda Y. A successful case of spastic paresis in a calf by partial tibial neurectomy. Nippon Juigaku Zosshi. 1989;51:213-214.

20. Sindou M, Fischer G, Goutelle A1 Schott B, Mansuy L Selective posterior rhizotomy in the treatment of spasticity. Rev Neurol (Ports). 1974;130:201-216.

21. Fasano VA, Barolat-Romano G, Ivaldi A, Squazzi A. La radicotoraie postérieure fonctionnelle dans le traitement de la spasticite cerebrale. Neurochirurgie. 1976;22:23-34.

21. Fasano VA, Broggi G, Zeme S. Intraoperative electrical stimulation for functional posterior rhizotomy. Scnnd7 Kehobil Med Suppi. 1988;17:149-154.

22. Dimitrijevic M. Spasticity. In: Swash MKC, ed. Scientific Basis af Clinical Neurology. Edinburgh: Churchill Livingstone: 1985.-108-115.

23. Lundberg A. Convergence of excitatory and inhibitory action on interneurons in the spinal cord. In: Brazier M, ed. The Incrmeuron. Los Angeles, Calif: University of California Press; 1969:231-265.

24. Raytner W. Spinal mechanisms for control of muscle length and tension. In: Davidoff R, ed. Handbook of the Spinal Cord. Vols 2 and 3. New York, Basel: Marcel Dekker; 1984:609-646.

25. Peter JC, Arens LJ. Selective posterior lumbosacral rhizotomy for die management of cerebral palsy spasticity. A 10-year experience. S Afr Med J. 1993;83:745-747.

26. Cahan LD, Adams JM, Perry J, Beeler LM. Instrumented gait analysis after selective dorsal rhizotomy. Dev Med Chad Neurol. 1990;32:1037-1043.

27. Park TS, Gaffney PE, Kaufman BA, Molleston MC. Selective lumbosacral dorsal rhizotomy immediately caudal to the conus medullaris for cerebral palsy spasticity. Neurosurgery. 1993;33:929-933; discussion 933-4.

28. Cohen A, Webster H. How selective is selective posterior rhizotomy. Surg Neurol. 1991;35:267-272.

29. Peacock WJ1 Staudt LA. Functional outcomes following selective posterior rhizotomy in children with cerebral palsy. ) Neurosurg. 1991;4:380-385.

30. Abbott R1 Johann-Murphy M, Shiminski-Maher T1 et al. Selective dorsal rhizotomy: outcome and complications in treating spastic cerebral palsy. Neurosurgery. 1993;33:851-857; discussion 857.

31. Nishida T, Thatcher S, Marty G. Selective posterior rhizotomy for children with cerebral palsy: a 7-year experience. Childs Nerv Syst. 1995;11:374-380.

32. Berman B, Vaughan C, Peacock W. The effect of rhizotomy on movement in patients with cetebral palsy. Am J Occup Ther. 1990;44:51 1-516.

33. Boscarino L1 Ounpuu S, Davis R, Gage ], DeLuca P, Effects of selective dorsal rhizotomy on gait in children with cerebral palsy. J Pediatr Orthop. 1993;13:174-179.

34. Dudgeon B1 Libby A, McLaughlin J, Hays R1 Bjomson K, Roberts T. Prospective measurement of functional changes after selective dorsal rhizotomy. Arch Phys Med Rehobil. 1994;75:46-53.

35. Bloom KX, Nazar GB. Functional assessment following selective posterior rhizotomy in spastic cerebral palsy. Childs Nerv Syst. 1994;10:84-86.

36. Kinghorn J. Upper extremity functional changes following selective posterior rhizotomy in children with cerebral palsy. Am J Occup Ther. 1992;46:502-507.

37. Euler A, Byrne W, Arment M, et al. Recurrent pulmonary disease in children: a complication of gastroesophageal reflux. Pediatrics. 1979;63:47-51.

38. GaskiII SJ, Willems K, Marlin AE. Selective posterior rhizotomy to treat spasticity associated with cerebral palsy: a critical review. Tex Med. 1992;88:68-71.

39. Laitinen LV, Nilsson S, Fugl-Meyer AR. Selective posterior rhizotomy for treatment of spasticity.; Neurosurg. 1983;58:895-899.

40. Xu L, Hong Y, Wang AQ, Wang ZX, Tang T. Hyperselective posterior rhizotomy in treatment of spasticity of paralytic limbs. Chin Med] (Engl). 1993;106:671-673.

41. McNeal D, Hawrrey C, Wolraich M, Mapel J. Symptomatic neurogenic bladder in a cerebral-palsy population. Dev Med Child Neurol. 1983;25:612-616.

42. Peter JC Hodman EB, Arens LJ, Peacock WJ. Incidence of spinal deformity in children after multiple level laminectomy for selective posterior rhizotomy. Childs Nerv Syst. 1990;6:30-32.

43. Abbott R. Complications with selective posterior rhizotomy. Pediatr Neurosurg. 1992;18:43-47.

44. Schi/man E, Erro MG, Mearte NV. Selective posterior rhizotomy: experience of 30 cases. Childs Nerv Syst. 1993;9:474-477.

45. Penn RD, Kroin JS. Intrathecal baclofen alleviates spinal cord spasticity. Lancet. 1984:1:1078. Letter.

46. Price G, Kelly J, Bowery N. The location of GABE-B receptor binding sites in mammalian spinal cord. Synapse. 1987;1:530-538.

47. Albright A. Intrathecal baclofen in cerebral palsy movement disorders. } Child Neurol. 1996;(11, suppl 1):S29-S35.

48. Mullet H. Treatment of severe spasticity: results of a multicenter trial conducted in Germany involving the intrathecal infusion of baclofen by an implantable pump delivery system. Dev Med Child Neurol. 1992;34:739-745.

49. Albright AL, Barron WB, Fasick MP, Polinko P, Janosky J. Continuous intrathecal baclofen infusion for spasticity of cerebral origin. JAMA. 1993;270:2475-2477.

50. Campbell S, Almeida G, Perm R, Coreos D. The effects of intrathecally administered baclofen on function in patients with spasticity. Phys Ther. 1995;75:352-362.

51. Bleck E. Orthopedic Management m Cerebral Palsy. Oxford: Blackwell Scientific; 1980.

52. Cosgrove A, Corry I, Graham H. Botulinum toxin in the management of the lower limb in cerebral palsy. Dev Med Child Neurol. 1994;36:386-396.

53. Silver C1 Simon S, Litchman H, The use and abuse of obturator neurectomy. Dev Med Child Neurol. 1966;8:203-205.

54. Bleck EE. The hip in cerebral palsy. Orthop CKn North Am. 1980:1 1:79-104.

55. Wheeler M, Weinstein S. Adductor tendonotomy-obturator neurectomy. J Pediatr Orthop. 1984;4:48-51.

56. Mertens P, Sindou M. Selective peripheral neurotomies for the treatment of spasticity. In: Sindou M. Abbott R, Keravel Y1 eds. Neurosurgery for Spasticity: a Mulridisciplmary Approach. Wien: Springer- Verlag; 1991:119-132.

TABLE

Ash worth Scale of Muscle Tone

10.3928/0090-4481-19971001-05

Sign up to receive

Journal E-contents