Cerebral Palsy

II. Abstract

Cerebral palsy (CP) is the most common cause of chronic neurological impairment in children. CP denotes a spectrum of neurological disorders arising from a lesion to brain areas involved in the initiation of movement, and clinical manifestation vary with the severity of the underlying insult. It is a syndrome of diverse etiologies and many potential risk factors, and the brain insult may occur pre-, peri-, or postnatally. While the fundamental motor impairment is reflected in abnormalities of muscle tone, which may range from minimal spasticity to devastating quadriplegia, many patients with CP will have additional findings, including epilepsy, mental retardation, speech and visual disturbance, gastrointestinal and feeding disorders, chronic pain, and others. Comprehensive management of motor disorders in CP is directed toward improvement of spasticity, dystonia, muscle stiffness, contractures and joint deformities, and other manifestations of poor motor control, such as sialorrhea. Treatment requires multidisciplinary care, adaptive devices, and symptomatic and supportive therapies. Pharmacologic management may warrant the use of combinations of medications with different mechanisms of action. Chemodenervating agents (phenol, alcohol, and botulinum toxin) and specialized orthopedic surgical procedures have become established treatment options to improve focal spasticity, dystonia, and other manifestations associated with CP. This educational activity will review pertinent information relevant to the etiology, epidemiology, clinical manifestations and management of CP, with emphasis on contemporary usage of botulinum neurotoxin.

III. Introduction

Cerebral palsy (CP) describes a group of permanent disorders that are attributed to non-progressive injuries to the developing brain. Abnormal muscle tone interferes with the development of movement and posture, causing activity limitations. The motor disorders of CP are often accompanied by disturbances of sensation, perception, cognition, communication, and behavior, by epilepsy, and by secondary musculoskeletal problems.¹

CP arises from a lesion to the brain and is an upper motor neuron disorder manifesting without associated anomalies of peripheral nerves or muscle.²

CP is the most common cause of chronic physical pediatric disability in Western societies. A fundamental characteristic of CP is that the lesion is static; CP is permanent yet nonprogressive at the brain level; however, clinical manifestations may change over time. Functional ability reflects the severity of the underlying cortical insult² and also may vary over time.³

IV. Epidemiology

Incidence and Prevalence

Estimates suggest that 764,000 children and adults in the United States manifest one or more symptoms of CP.⁴ Approximately 8,000 babies and infants and 1200-1500 preschool-age children are diagnosed with the condition each year.⁴ The prevalence of CP has risen over a 40-year period, from approximately 1.5 per 1000 live births in the 1960s to about 2.5 per 1000 in the 1990s. Over the past two decades, three surveillance studies on children with CP in the United States have been performed. Using a multiple-source case identification method, in 1992 the Metropolitan Atlanta Developmental Disabilities Study by Yeargin-Allsopp et al assessed the prevalence of mental retardation, CP, hearing impairment, and visual impairment among children who were 10 years of age between 1985 and 1987 in the five Georgia counties.⁵ This study found the prevalence of CP in this geographical area to be 2 per 1000 children. A population-based cohort study in 1993 by Cummins et al covered children born between 1983 and 1985 in four northern California counties to examine the impact of demographic shifts and changes in perinatal medicine on the distribution of CP.⁶ In 2008, the surveillance study of 8-year-old children with CP by Yeargin-Allsopp et al addressed specific regions within three states that established prevalence of CP in the United States as 3.6 per 1000, comparable to CP prevalence throughout the world.⁷ This study brings attention to an increase in the more disadvantaged urban populations, since incidence was highest in the inner-city areas. A recent study by Wu et al suggests that though there is increased representation for the African-American population in California for low weight, 29% higher low-birth-weight rate, the risk of CP for that birth-weight group was lower for the African-American population.⁸ The increased incidence of prematurity and risk for CP in a disadvantaged population may not be unique to the United States.⁹

Over the past 15 years, the proportion of low-birth-weight infants among all children with CP has risen, and newborns weighing less than 2500 g now represent half of all cases of CP and just over half of the most severe cases.¹⁰ Lately attention has been placed on late preterm children as a group to follow carefully for increased medical and developmental disorders.¹¹.

Economic Costs

Economic costs associated with CP are considerable. Estimated lifetime expenditures for people born with CP in 2000 were $11.5 billion in 2003 dollars, representing a per-person cost of $921,000. Total direct costs for people with CP born in this single calendar year were estimated at $2.2 billion, emphasizing that the majority of fiscal consequences relate to indirect losses sustained by households and workplaces when individuals with CP cannot work, are limited in the type of work they can perform, limit the economic productivity of family caregivers, or die prematurely.¹² Calculations for costs from other countries yield similar expenditures.^13-15

Additional Impairment-Associated Conditions

The potential for musculoskeletal deformity is present in all children with CP and is the reason many parents seek treatment for the child with spasticity and other tone problems that interfere with motor skills.¹⁶ In particular, the progressive nature of the effect of hypertonia on deformity is a compelling reason for aggressive treatment of the muscle tone.

The majority of patients with CP have additional impairments, with estimates ranging from 25% to 80% depending on the subgroup. Epilepsy is present in 20% to 40%, and is most common among hemi- and tetraplegics. Many patients with CP have a degree of cognitive impairment, varying with the type of CP and especially frequent in cases associated with epilepsy.¹⁰ Learning disabilities and working memory deficits frequently confound educational efforts.^17-21 About 20% to 50% of children with CP may have some degree of speech impairment.^22,23 Visual perceptual impairments and low visual acuity is reported in almost three-quarters of children; strabismus is an especially common ophthalmologic finding. Primary urinary incontinence occurs in 23% of children and adolescents with CP. Half of all children with CP have gastrointestinal and feeding problems, including most prominently choking, vomiting, prolonged feeding, and in severe cases a requirement for parenteral administration. Chronic pain is reported by more than 25% of adults. Overall, half of all people with CP require an assistive device to maintain mobility.²⁴ It stands to reason that the more severe the motor impairment, the greater need and reliance on equipment for mobility.²⁵

V. Clinical Features

Classification

CP has traditionally been categorized using different schemes based on muscle tone, anatomic distribution, severity, or some combination of these indicators.

Figure 1. Classification of CP. CP may be classified by a variety of methods. The site of the underlying, static lesion determines the affected limbs and laterality of the motor impairment. The degree of clinical impairment in CP ranges from mild uncoordination to complete dependency, which in turn has a great effect on overall functional status and prognosis. Reprinted from Berker and Yalcin, permission pending.²⁶

In one common scheme, CP is classified into four principal forms according to the predominant type of motor disorder:

• Spastic cerebral palsy is the most common form, affecting 70% to 80% of patients. It is an expression of an upper motor neuron syndrome with hyperreflexia, clonus, extensor plantar responses, and primitive reflexes.^26,27 Muscle hypertonia in conjunction with the growth of the affected child may lead to fixed contractures, torsional deformities of long bones, and joint instability, further impeding motor performance.²⁸ Affected muscle groups may be localized mainly to lower extremities with variable upper-extremity involvement (diplegia), to three extremities (triplegia), to all extremities (quadriplegia), or to one side (hemiplegia).⁴

• Dyskinetic cerebral palsy affects 10% to 20% of patients,⁴ and is characterized by uncontrolled, excessive movements of the extremities. Dyskinetic CP can encompass athetosis, dystonia, chorea, and ballismus. The muscles of facial expression and the tongue are commonly involved, leading to grimacing and drooling.^4,29

• Ataxic cerebral palsy affects 5% to 10% of patients, and typically manifests as impaired balance and depth perception. Individuals with ataxic CP have a characteristic wide-based gait, poor coordination, and difficulty with precise movements, and may exhibit intention tremor.⁴

• Mixed forms of CP are not unusual. The most common mixed form includes spasticity and athetoid movements.⁴

A common language on classifying CP has been developed for the European registry by the SCPE (Surveillance of Cerebral Palsy in Europe) working group; they classify the disorders by laterality and tone. 53.9% of the CP children had a bilateral spastic CP, 31.0% had unilateral spastic CP, 6.6% were dyskinetic, and 4.1% ataxic.³⁰

The Gross Motor Functional Classification System (GMFCS) is also widely used as a tool for assessing basic motor capability. The GMFCS is especially useful for research, teaching, and administration, as functional levels remain fairly stable over time. This system, which is based on self-initiated movement with particular emphasis on sitting (reflecting truncal control) and walking, encompasses five categories³¹:

*Ambulation descriptors are a primary determinant of functional level. Sitting descriptors vary with age group; descriptors shown here relate to observations made before the child’s second birthday. Additional descriptors applicable to older age groups are specified by Palisano et al.³¹

There is a spectrum of motor difficulty for the child and later the adult with CP. The GMFCS and the manual ability classification systems allow us to categorize on the basis of function rather than impairment. For the majority of children and adults with CP, it appears that function is relatively stable over time.^3,32-40 Some children with CP require no special assistance for mobility or activities of daily living, while children with GMFCS IV or V need extensive care throughout their lives. Lifespan is near normal for the majority of children with CP and shortened for those with GMFCS IV or V that have a combination of dysphagia, seizures, and poor head control.^41-46 Although the brain damage underlying CP is nonprogressive, secondary conditions (such as muscle spasticity) may fluctuate over time and lead to a decline in function.^4,47,48 Strength is lost over time as well, which contributes to a decline in function.⁴⁹

VI. Pathophysiology and Pathogenesis

Etiology and Risk Factors

CP is a syndrome with diverse etiologies and numerous potential risk factors. The many potential etiologies of cerebral palsy are often described as encompassing the “four P’s”: prenatal, perinatal, postnatal, and prematurity. Prematurity is more properly considered a risk factor.

Table 1. Cerebral Palsy: Causes and Risk Factors.^4,54 The many potential etiologies of cerebral palsy are often described as encompassing the “four P’s”: prenatal, perinatal, postnatal, and prematurity. Prematurity is more properly considered a risk factor.

The brain insult may occur pre-, peri-, or postnatally.²⁴ Most children with CP are born with it (congenital CP), and for this population there are only a few key mechanisms by which permanent injury to cortical motor regions is presumed to occur. Periventricular leukomalacia arises from several causes, including maternal or fetal infection contracted during vulnerable periods in fetal development of the central nervous system, and leads to deteriorated white matter architecture characterized by actual cavitation and impaired signal propagation.

Figure 2. Periventricular leukomalacia. White-matter injury in a newborn. Arrows in panels A and B indicate enhancing lesions throughout the periventricular white matter of both cerebral hemispheres. Hypodensities in panel C represent cavitation at sites of coagulation necrosis; these areas are visualized as hyperintense regions on T2-weighting (panel D). Reprinted from Back, permission pending.⁵⁰

Cerebral dysgenesis results from myriad possible causes, including genetic mutation, and leads to malformations of relevant areas. Fetal or neonatal cerebrovascular accident (stroke and intracranial hemorrhage) leads to ischemia of motor sites. Finally, hypoxic-ischemic encephalopathy (also referred to as intrapartum asphyxia) destroys tissue in the motor cortex and other areas of the brain.

These mechanisms should be viewed as representing final pathways leading to congenital lesions of the motor cortex; they are not inherently “causes” of CP. What remains to be established more clearly is the relationship between these root mechanisms of cerebral damage and the many maternal, infectious, and other risk factors clearly established to be associated with CP. Intense interest is currently focused on the role of inflammation arising within the intrauterine milieu as a linking and contributing factor to white-matter damage or arrested brain development at critical periods in fetal development.^50-53

Figure 3. Inflammatory processes in perinatal brain injury. Intracerebral inflammation elicited by provocative factors in the fetal/maternal environment may play a role in white-matter damage and arrested brain development. Illustrated are two common insults associated with CP: infection, represented by contact with lipopolysaccharide derived from bacterial cell wall, and hypoxia/ischemia. Proinflammatory cytokines participate in a cascade of events that may be either neurotoxic (leading to cell damage or cell death) or neurotrophic (encouraging proliferation and survival). Mechanisms favoring the predominance of one outcome or the other remain undefined. Abbreviations: ATP, adenosine triphosphate; BDNF, brain-derived neurotrophic factor; H/I, hypoxia ischemia; IL, interleukin; LPS, lipopolysaccharide; MAPK, mitogen-activated protein kinase; MCP, monocyte chemotactic protein; MMP, matrix metalloproteinase; MyD88, myeloid differentiation primary response protein 88; NFκB, nuclear factor κB; NGF, nerve growth factor; p, protein; R, receptor; RA, receptor antagonist; RAcP, receptor accessory protein; ROS, reactive oxygen species; TNF, tumor necrosis factor. Reprinted from Girard et al., permission pending.⁵³

Acquired CP is infrequent and can often be traced to a specific etiology, such as meningitis or encephalitis, or brain damage subsequent to accidental injury, motor vehicle accident, or child abuse. While complications of labor and delivery were once thought responsible for a substantial proportion of CP, there is now consensus that birth complications cause only 5% to 10% of cases.⁵⁴

VII. Diagnosis

CP is a clinical diagnosis that is established based on findings of delayed motor milestones (the most common concern prompting medical evaluation), abnormal muscle tone, atypical movement, and hyperreflexia. A careful history should confirm that the child is not losing function, thus excluding the presence of a progressive disease. CP may not be detected for months or years after birth; however, the diagnosis is usually considered in a majority of affected children by 2 years of age. Infrequently, diagnosis will be delayed into middle childhood or early adolescence.^2,24

Though laboratory tests are not necessary for the diagnosis if the history is consistent with high-risk babies, imaging is recommended for children with the clinical diagnosis of CP. Magnetic resonance imaging (MRI) is the imaging modality of choice in the diagnostic evaluation of children presumed to have CP, and should be obtained if the diagnosis is uncertain; it is especially sensitive for identifying periventricular leukomalacia. MRI, and to a lesser extent computed tomography (CT), may be valuable for identifying a small number of conditions that mimic CP but are treatable.²⁴

VIII. Treatment

The pattern of disability, deformity, comorbid medical conditions, and overall quality of life associated with CP is unique to every patient, and management decisions require careful consideration of potential ramifications across a spectrum of individual impairment. The consequences of CP range from those that are fundamental and clearly apparent, such as seizure, mental retardation, and pervasive spasticity, to those that are less obvious but perhaps especially poignant, as when excessive drooling mandates frequent daily clothing changes or damages adaptive devices and keyboard electronics that an individual requires to interact with others.⁵⁵

Multidisciplinary care is therefore the foundation of CP management, incorporating varied specialist input to identify specific impairments and needs, and formulate an appropriate, comprehensive plan that addresses core disabilities. The aggressive use of multidisciplinary symptomatic and supportive therapies employed in conjunction with adaptive devices represents the most effective strategy for reducing disability and improving quality of life.

Motor disorders in CP mandate a focus on spasticity, dystonia, muscle stiffness, contracture, joint deformity, muscle weakness, and other aspects of abnormal motor control. Spasticity plays a key role in the development of deformity and, in conjunction with imbalance and restricted weight bearing, initiates a chain of events that may lead to fixed contracture, joint instability, and possibly degenerative arthritis; thus, the management of spasticity is very important in a growing child.⁵⁶

Consistent with a broad-based approach, pharmacologic management of the CP patient may necessitate the use of combinations of medications with different mechanisms of action⁵⁷; importantly, administration of multiple medications requires thorough assessment of the potential for adverse interactions, judicious dosage selection, and careful surveillance of clinical response. In young children with CP, management of spasticity or dystonia with oral medications such as diazepam and baclofen raises concerns because of safety and tolerability issues associated with long-term use. These and other oral medications may be useful for treating muscle hypertonicity in older subjects and forestalling the development of the fixed deformities and postures. Chemodenervating agents (phenol, alcohol, and botulinum toxin) and specialized orthopedic surgical procedures have become established additions to contemporary treatment regimens to improve focal spasticity or dystonia associated with CP.

An integrative graph has been developed that illustrates principles of common interdisciplinary treatment modalities focused on improving gross motor function.⁵⁸

Role of Neurotoxins

Botulinum neurotoxin (BoNT) inhibits the release of acetylcholine from the presynaptic nerve terminal, thereby causing local muscular weakness. Injection of BoNT reduces muscle tightness, and leads to improved movement and function of treated limbs. BoNT has been used to improve a range of upper and lower impairments in CP, including spastic equinus, adductor spasticity, hamstring spasticity, crouched gait, seating difficulties, spastic upper limb postures, and others.^28,59

Seven serotypes of BoNT (A, B, C, D, E, F, and G) are known. OnabotulinumtoxinA (Botox^®), abobotulinumtoxinA (Dysport^®), incobotulinumtoxinA (Xeomin^®), and rimabotulinumtoxinB (Myobloc^®/Neurobloc^®) are different BoNT products that are commercially available; although none of these is FDA approved for the treatment of children with CP, onabotulinumtoxinA is FDA approved for the treatment of upper limb spasticity in adults over the age of 18.

There is accumulating clinical evidence from small series and systematic reviews to suggest that BoNT can be a useful component of a multidisciplinary approach to the management of CP, with data confirming the efficacy of BoNT for the management of both upper- and lower-extremity spasticity in appropriate patients. A 2010 Cochrane review encompassing 10 trials of upper-limb spasticity in children with CP determined that the combination of BoNT-A and occupational therapy (OT) is more effective than OT alone in attaining improvements in activity level and goal achievement, but not for improving quality of life or perceived self-competence. This review found high-level evidence to support the use of BoNT-A as an adjunct to OT for managing upper-limb spasticity, but found no evidence to support its use as a sole therapy.⁶⁰ This latter caution is consistent with European expert guidelines compiled from pooled data for 10,000 patients across a range of common CP interventions, which described the BoNT formulations generally as a major therapeutic intervention for children with CP, but one which should never be employed in isolation.⁶¹ The safety of BoNT-A was confirmed in a meta-analysis of 20 trials encompassing 882 patients; while adverse events were found to be more common in children with CP than with other conditions, good safety during the initial months of therapy was demonstrated, and six studies reported no adverse effects.⁶² The superiority of BoNT injections over placebo for management of equinus deformity of the foot in CP was confirmed in a meta-analysis of trials in which outcome was assessed by physician rating, by video analysis, and by subjective analysis of well-being; adverse events were mild and transient.⁶³

The ideal candidate for BoNT injection is one with marked spasticity but without fixed contracture. If the main impediment to movement is poor coordination rather than hypertonic resistance, botulinum injections are unlikely to improve function. Fixed contracture similarly diminishes the potential effect.^{64 65}

Considerations for BoNT Therapy

1. The muscle

Muscle structure and architecture in a child with CP has been studied and found to be altered in fiber type, due to chronic stimulation.^66,67 Muscle weakness is due to reduced central drive, possible abnormal neural maturation, insufficient and disorganized motor recruitment, impaired voluntary control and impaired reciprocal inhibition, altered setting of muscle spindles, and reinforcement of abnormal neural circuits. There is also alteration of muscle tissue with selective atrophy of fast fibers and altered myosin expression, changes in fiber length and cross-sectional area, changes in the length-tension curve, reduced elasticity, and impoverished muscle-tissue development.⁶⁸ Technological advances in study method such as microdissection of the individual muscle fiber allows a closer look at the sarcomere of the spastic muscle.^69-72 Dr. Rich Lieber’s elegant studies reveal that, although the isolated cells of spastic muscles may be stiffer compared to normal cells, bundles from spastic muscle are actually less stiff compared to normal muscle fiber bundles because the extracellular matrix from spastic cells has inferior material properties. The extracellular matrix accounts for the “feel” of the muscle, which experienced clinicians or surgeons can detect in exam. ⁷²

Ultrasound is another way to examine the properties of muscle and this provides insight to the diameter-strength relationship as well as the tendon-muscle relationships.^73-75

The first injector of botulinum toxin in the United States, Dr. Andre Koman, has shifted his study to the cellular level of understanding of the effects of botulinum toxin on the muscle. Following denervation, a sequence of cellular events eventually leads to neuromuscular junction stabilization, remodeling, and myogenesis, and the recovery of muscle function. There are two stages to the recovery, aneural and neural, and the insulin-like growth factor-1 signaling pathway plays a central role in the process.⁷⁶ In his series of rat studies, he and his colleagues have shown that the effect of the drug and the intracellular events correspond well to the human clinical effects of neurotoxin injections.⁷⁷ Recent use of MRI demonstrates that the atrophic effect to the muscle following botulinum toxin in healthy volunteers may persist for longer than the clinical effect.⁷⁸ A brief but important letter to the editor demonstrates residual evidence of damage to the muscle in children with CP after the clinical effects have diminished.⁷⁹ The major muscle groups and gross localization for the injection sites are illustrated in Figure 4.

Figure 4. Common injection sites and potential target muscles in spasticity treatment.¹²⁷

2. The Injection

Technological adjuncts to injections (such as sonography) make possible very precise administration to target muscle bellies, minimizing the risk of subcutaneous, intravascular, or too-deep injection in young children with atrophied musculature.^80-84 Studies suggest that there is inaccuracy using anatomic localization, so targeted localization of the muscle is warranted.⁸¹ Not only is the localization of the muscle necessary but localizing motor endplates is possible and may be beneficial. Location of motor endplates have been mapped and provide a guide for targeted injection.^80,85,86 This has the advantage of maximizing the toxin load to the motor endplate area of muscle; however, functional outcomes have not been correlated with injection technique. Pros and cons of injection localization seem to favor the use of ultrasound. In particular, ultrasound does not increase anxiety, which should be minimized for children. Though it is impractical to obtain MRIs to determine area of last injection, an effort to change injection sites has the potential to improve subsequent injections, since the muscle may not have had a chance to recover if injected into same site. Sonoelastography is an ultrasound technique that allows the assessment of tissue elasticity. Its application in the musculoskeletal field includes the evaluation of muscle contraction status. On sonoelastographic images, a relaxed muscle structure will appear mostly soft (green- yellow-red), while contracted or degenerated muscle fiber will appear hard (blue).⁸⁷

A recent report utilizing sonoelastrographs suggests this technique may be a less invasive way to monitor the effects of toxin over time on the muscle and determine injection site.⁸⁸

Dilution is another factor in enhancing uptake and effect of the toxin. Multiple studies have demonstrated that larger volumes injected will better distribute the toxin.^85,89,90

The effects of injection also vary with the extent of sedation. The overriding principle is to first do no harm, so a frank discussion with parents to determine the level of sedation necessary to obtain good localization and good experiences for the family is wise. The gamut of choices from nitrous to versed to EMLA (lidocaine and prilocaine) cream and distraction have all been utilized and should be available if necessary to achieve good results.^91,92 Children and parents will have different reactions and concerns, especially with recent acknowledgment of the risks of anesthesia during frequent minor procedures.⁹³

3. Outcomes

Most of the studies that favor localization and dilution rightfully evaluate the body-structure aspects of outcome measures such as the modified Ashworth scale, strength, and change in range. Function and change in participation is much harder to connect to the issues of injection technique, and may need to be evaluated over a longer time course in children.

Although BoNT treatment may improve function in the short term, it has the potential in the growing child to affect long-term muscle growth and function. A considered and comprehensive intervention approach to the use of BoNT, with recourse to objective outcome measures within a specialized multidisciplinary setting, is therefore recommended, particularly in ambulatory children, until further evidence is available on the long-term outcome of early injections in young subjects with CP.⁹⁴ Outcome measures should reflect the International Classification of Functioning paradigm of body structure and function, activities, and participation.^95,96 The majority of studies focus on impairment ratings and outcomes such as reduction of spasticity and improved range of motion, which is the direct effect of the injection.⁹⁷

Realistic expectation and a set goal is expected to favor a better perceived outcome.

IX. Upper-Limb Spasticity

There is high-level evidence to guide dosage selection and to support the safety and effectiveness of BoNT in combination with OT for management of upper-limb spasticity in children with CP. This evidence, collated in a recent systematic review, encompasses randomized published and unpublished studies comparing BoNT with or without OT versus placebo/no treatment, BoNT plus OT versus OT alone, BoNT plus OT versus BoNT alone, and BoNT alone versus OT alone.⁶⁰

OnabotulinumtoxinA plus OT were compared against OT alone in a randomized, controlled study enrolling 43 children aged 3 to 16 years (mean, 8.6 years) with passive joint range of motion meeting prespecified limits (elbow extension to neutral, wrist extension to 30° past neutral with fingers extended, supination of the forearm of 30° past neutral, and thumb extension to neutral), ability to initiate movement of the fingers, and a Modified Ashworth Scale spasticity score of 2 or higher at the elbow or wrist. Children randomized to BoNT received individualized injections of neurotoxin under general anesthesia; injections were under muscle-stimulator-assisted guidance to improve accuracy. All subjects received weekly OT (1 h session) for 4 weeks. The total injected dose of onabotulinumtoxinA did not exceed 12 U/kg, to a maximum of 300 U (onabotulinumtoxinA­­­­). Objective outcome measures were assessed at baseline, 3 months, and 6 months, and patients were also asked to describe whether they felt better, worse, or unchanged with regard to function and cosmesis. At 3 months, children allocated to receive the intervention performed significantly better in terms of body structure and activities participation, while reporting improved self-perception. Elbow and wrist tone and spasticity improved significantly versus control subjects. At 6 months, differences between intervention and control groups persisted for the measures of body structure, but not for activities participation or self-perception. The most common adverse events were post-anesthesia vomiting and cough (n=4), as well as excessive weakness in the injected limb (n=5).⁹⁸

OnabotulinumtoxinA injections to the upper limb in 10 children receiving a relatively intensive occupational and physical therapy program achieved a beneficial effect on wrist flexor tone and self-reported improvements in core tasks of daily function in a small, randomized study.⁹⁹ The effect was present as early as 2 weeks after injection and remained for 9 months. Dorsal flexion of the active wrist showed a trend toward improvement that endured for at least 9 months, although this effect did not reach statistical significance. Nine of the 10 children receiving onabotulinumtoxinA reached their treatment goals, which varied from use of a knife, tying shoelaces, and carrying cups to climbing, shaking hands, and hairdressing. Children in the onabotulinumtoxinA group were clearly more satisfied with their treatment result than controls in this non-blinded study, although the judgment of parents from both groups did not differ greatly.⁹⁹

An early randomized, controlled, single-blind trial evaluated onabotulinumtoxinA injections at a dosage of 2-6 U/kg into at least one of three muscle groups (biceps, volar forearm, or adductor pollicis) in 14 children with hemiplegic CP (mean age, 5.6 years). Location was guided by muscle palpation and empiric anatomic assessment. Biceps and volar flexors received injections at two sites, while the adductor pollicis and pronator teres were injected at one site only. Children in control and treatment groups continued with community-based OT. The primary outcome measure was Quality of Upper Extremities Test (QUEST) score, an objective standardized measure given as a percentage of 100, reflecting the quality of upper-extremity function through multiple domains. A clinically and statistically significant improvement in quality of function was found at 1 month, with QUEST scores for treated children rising from 19.2% at baseline to 32.5%, compared with a 1.7% change in the control group. Moderate improvements were maintained up to 6 months after injection. Treatment was well tolerated, with only one child reporting transiently decreased grip strength.¹⁰⁰

A notable cosmetic benefit of reduced involuntary elbow flexion was reported along with significant improvements in maximum active elbow and thumb extension and significantly reduced muscle in a true double-blind trial (including sham injections) enrolling 14 patients (mean age, 9 years). All placebo patients later requested and received onabotulinumtoxinA treatment. Injections were guided by empiric physical examination. No patient received OT during the course of the study. Improvements in elbow and wrist tone and muscle extension were apparent by 2 weeks, and were still apparent by objective and subjective assessment at 12 weeks. While hand grasp-and-release scores improved, fine motor function did not and in some cases deteriorated.⁶⁴

The importance of multimodality therapy in the management of spasticity was confirmed most recently in a randomized, controlled trial enrolling 80 children with spastic quadriplegia, triplegia, or hemiplegia. Children were randomized to onabotulinumtoxinA plus OT, onabotulinumtoxinA alone, OT alone, or no treatment (control). Patients randomized to treatment received a single session of injections and 12 weeks of OT. The combination of onabotulinumtoxinA and OT resulted in accelerated attainment of functional goals measured by multiple standardized assessment tools. The onabotulinumtoxinA plus OT group had the largest change from baseline at both 3 and 6 months and was the only group to have a clinically significant change of at least two grading points at both follow-up assessments. This was also the only group whose members on average achieved their prespecified functional goals at 3 months, most of which were leisure related or self help (e.g., pushing an arm through a sleeve while dressing). All groups except the control group on average achieved their goals at 6 months. There were no significant differences between groups receiving OT alone or onabotulinumtoxinA alone. Patients treated with onabotulinumtoxinA alone and in particular those patients treated with onabotulinumtoxinA plus OT demonstrated improvements in active supination, an important and promising outcome for enhanced function. There was a significant reduction in muscle tone among all treated patients within 2 weeks of injection, which returned to baseline by 6 months. Adverse events were generally mild, and no instances of excessive weakness were reported.¹⁰¹

The impact of onabotulinumtoxinA dosage on functional improvement continues to attract interest focused on defining optimal recommendations. Children receiving low-dose, high-concentration onabotulinumtoxinA injections plus OT demonstrated better movement quality and better function in comparison with children receiving OT alone in a randomized trial enrolling 42 children with a mean age of 4 years (range, 2-8 years). All children were GMFCS level I. Injection of concentrated onabotulinumtoxinA (maximum, 8 U per kg; 0.5 ml dilution) was assisted by electromyography and muscle stimulation. Children in the treatment group attained better quality of upper-limb movement sooner than patients receiving OT alone. A single session of dual-mode localized injection produced greater 1-month and 3-month gains in upper-limb quality of movement and function, as well as 6-month gains in function. There were no adverse events attributed to BoNT.¹⁰²

More recently, the effects of high and low doses of onabotulinumtoxinA on upper-extremity function were compared in a study enrolling young children (mean age 6 years, 2 months) with spastic hemi- or triplegia. The high-dose group received onabotulinumtoxinA at 3 U/kg (common flexor), 2 U/kg (biceps), 1.5 U/kg (brachioradialis or pronator teres), or 0.6 U/kg (adductor/opponens pollicis) to a maximum of 20 U. The low-dose group received 50% of this dosage, diluted to maintain a comparable volume. Three investigators jointly determined which muscles to inject on the basis of grasp activity. Injection locations were determined empirically. All children received OT. Improvements at 1 and 3 months were comparable between groups, supporting the use of a lower dose in appropriate settings and emphasizing the importance of volume and multiple injections over mass of onabotulinumtoxinA for clinically meaningful delivery of active agent to target sites. The authors noted a potential cost saving in their study of $180 (Canadian) per child, and suggested that lower doses per injection potentially support strategies targeting additional muscle groups before a maximal dosage is reached.¹⁰³

X. Lower-Limb Spasticity

BoNT is a key adjunct for management of lower-limb spasticity, and is especially useful for reducing excessive gastrocnemius/soleus tone associated with equinus foot deformity. In the first large, randomized, double-blind, placebo-controlled trial to demonstrate gait improvements following BoNT, 56 hemi- and diplegic children with dynamic equinus foot deformity received single injections of onabotulinumtoxinA into the medial and lateral gastrocnemius muscles. All patients continued prestudy physical therapy regimens and, in some cases, orthotics use. Patients receiving onabotulinumtoxinA demonstrated significant improvements in standardized gait scores and in individual components of gait pattern and ankle position, with 61% of subjects achieving improvements in gait pattern at 8 weeks. Hemiplegic and diplegic subjects attained similar improvements. No serious adverse events were reported, and the majority of adverse events experienced by patients were mild.¹⁰⁴

In a follow-up to this single-treatment investigation, a large, open-label, multicenter clinical trial evaluated the long-term safety and efficacy of repeated onabotulinumtoxinA injections administered as an adjunct for management of equinus gait. Of 207 enrolled patients, 155 (75%) completed at least 1 year of treatment. Forty-two patients completed 2 years or more and 7 patients completed 3 years or more. The mean duration of onabotulinumtoxinA exposure was 1.46 years per patient for a total of 302 patient-years. The average onabotulinumtoxinA dose was 84.8 U per injection session. One hundred fifty patients received four treatments, and 31 patients received eight treatments. Responses were maintained in 41% to 58% of patients for 2 years, with gait pattern and ankle position demonstrating improvements at every visit. The most common adverse events were stumbling, leg weakness, leg cramps, and calf atrophy, which occurred in 1% to 11% of patients.¹⁰⁵

Using video gait-analysis scoring with an expanded scale for assessing toe-walking, investigators documented improvements in walking pattern in children with CP following a single injection of abobotulinumtoxinA to the gastrocnemius. Fifty percent of children treated with abobotulinumtoxinA showed significant clinical improvement. Improvement was maintained to 12 weeks, and was superior to that obtained with physical therapy and splinting alone.¹⁰⁶

Adductor spasticity impairs walking, sitting, standing, other movements, and personal hygiene, and is an important risk factor for hip dislocation. In a double-blind, randomized, placebo-controlled study, a single session of abobotulinumtoxinA administered to 33 children (including 16 with a GMFCS ≥4) improved knee-knee distance significantly at 4 weeks, and was also associated with improvements in goal attainment. Improvements in knee-knee distance remained apparent at 12 weeks. The most commonly observed adverse events were muscular weakness, dysphagia, and urinary incontinence.¹⁰⁷

Surgical interventions to improve gait are typically considered only after attainment of a stable gait pattern, as premature operation is associated with a higher risk of failure and unpredictable results. Three-dimensional gait analysis and botulinum injections, in conjunction with active motor training and appropriate orthotic management, delayed the progression to surgical intervention in a retrospective review of 424 children with CP. The patients were stratified by management pattern: Group 1 comprised 122 patients managed according to best-practice guidelines in orthopedics, Group 2 included 170 patients managed per best-practice guidelines, in addition to gait analysis, while Group 3 comprised best practice in addition to gait analysis and botulinum injection. One hundred thirty-two patients who had gait analysis also received BoNT injections. There was a significant decrease in the prevalence of orthopedic surgical procedures among children in Group 3 compared with Group 1 (p < .00001 at 4 to 8 years of age) and Group 2 (p < .0025 at 4 to 9 years).¹⁰⁸ While retrospective, these results suggest that the treatment of primary motor problems with BoNT extended the duration of time within which a stable gait pattern may emerge, facilitating a more appropriately timed surgical assessment and reducing the likelihood of premature operative intervention.¹⁰⁹

XI. Sialorrhea

Sialorrhea (drooling) is a challenging problem in patients with CP, affecting care, oral health, quality of life, and social function. BoNTs, which have attracted interest for management of a range of hypersecretory disorders, block acetylcholine release at the neurosecretory junction following administration into the salivary gland and effectively reduce saliva production.^110-113 Reports from larger trials and a recent international consensus statement affirm that onabotulinumtoxinA and other BoNT formulations are useful selections for safe and effective management of drooling in carefully selected children with CP; note is made, however, that caution with this approach is warranted in the presence of dysphagia and CP with GMFCS level V.¹¹⁴

A controlled, open-label, clinical trial compared onabotulinumtoxinA therapy with the anticholinergic medication scopolamine in 45 children with CP. The first phase evaluated scopolamine applied as a transdermal patch, applied behind the ear and changed within every 72 hours until assessment on the tenth treatment day. After an appropriate washout period, a single weight-adjusted dose of onabotulinumtoxinA was injected under ultrasound guidance into the submandibular glands bilaterally. Drooling severity was measured at baseline, during application of scopolamine, and at different intervals after BoNT injections up to 24 weeks; semiquantitative measurements were obtained with the Drooling Quotient, the Teacher Drooling Scale, and visual analog scales.

Thirty-nine children completed the trial. Treatment with scopolamine and BoNT were each associated with an approximately 50% response rate and a clinically apparent, significant reduction in drooling. Although scopolamine has systemic availability and is thus likely to affect submandibular, parotid, and sublingual glands, the short-term effect of focal submandibular BoNT injection was of comparable magnitude, suggesting a strong anticholinergics effect and also indicating that submandibular glands produce a major portion of saliva. Maximal effect of BoNT treatment was seen 2 to 8 weeks after injection. BoNT injections were associated with fewer and less serious side effects than applications of scopolamine. Overall, the most common side effects across treatment groups were xerostomia, restlessness, somnolence, visual disturbance, and confusion.¹¹⁵

In a large, prospective, observational series of 131 patients treated for drooling with intra-submandibular onabotulinumtoxinA, investigators demonstrated an objective and subjective response rate of approximately 50%, similar to that found in smaller studies. Although this was a heterogeneously impaired population, over 90% of subjects had CP. Improvements were reflected in a significant mean reduction in Drooling Quotient from a baseline of 29 to 15 after 2 months and 19 after 8 months (p < .001). The mean visual analog scale score decreased from 80 to 53 after 2 months, increasing to 66 after 8 months (p < .001). Responders benefited from injection for a median of 22 weeks. After 33 weeks, 25% of initial responders (11.3% of the entire population) still showed a clinically significant response, with a very small number of patients experiencing continued drooling relief after 1 year. Secondary beneficial effects observed in a small number of patients following injection included improved oral hygiene (reduced perioral dermatitis or reduction in halitosis) in four patients (3.1%) and improved speech in another four patients. These secondary effects generally disappeared after 8 months. BoNT injections were usually well tolerated. The most common side effects of treatment were changes in saliva (thickening), transient difficulty in swallowing, transient deteriorated feeding behavior, and xerostomia.¹¹⁶

The safety and efficacy of abobotulinumtoxinA were evaluated in 24 children with CP randomized to placebo or to treatment with two intraparotid injections (100 U and 140 U 4 months later). The median frequency score (4), median severity score (5), and median of the total drooling score (9) were unchanged among patients receiving placebo. The median frequency of drooling score declined from 4 to 3 (p = 0.034), the median severity score declined from 5 to 4 (p = 0.026), and the median total score changed from 9 to 7 (p = 0.027) in the treatment group, with two patients in this group experiencing transient increases in drooling after the injection. Sixteen patients received the second injection (nine in the treatment group and seven in the placebo group), with additional improvement noted among five. Three patients who did not respond to the first injection responded to the second injection. Treatment was well tolerated.¹¹⁷

Most recently, three different doses of BoNT-B were evaluated for management of persistent sialorrhea in 27 children with CP. Patients were randomized to a low (1500 mouse units [MU]), medium (3000 MU), or high (5000 MU) dose of BoNT-B injected into the parotid and submandibular glands bilaterally, or to control treatment. All patients were assessed at baseline, 4 weeks, and 12 weeks. Results showed that a 3000 MU injection of BoNT-B significantly improved the frequency and severity of drooling, while the 1500 MU dose was ineffective and the 5000 MU dose yielded no additional benefit but more side effects in comparison with the 3000 MU dose. In all children receiving either medium- or high-dose BoNT-B, reduction of drooling lasted for 12 weeks, and in five children the effect was prolonged for up to 6 months after treatment. No serious adverse events were reported in the low-dose group. Side effects reported by the other dosage groups included dense saliva, difficulty swallowing, and xerostomia; these were more frequent in the high-dose group.¹¹⁸

Safety

Appropriate incorporation of BoNT into clinical management strategies for CP must be undertaken with full awareness of potential systemic adverse events. In contrast to established, typically transient localized treatment-emergent adverse events such as muscle weakness, the possibility of systemic adverse events warrants close attention to dosing guidelines for the various commercial preparations of BoNT, as well as careful and precise administration to target sites. Botulinum dose limits should be carefully reviewed for children with GMFCS levels of IV and V, as well as for those with histories of aspiration or comorbid respiratory disease. Alternatives to mask anesthesia should be considered.¹¹⁹ A recent study that evaluates the comparison of medical events pre- and then postinjection shows no increase in adverse events post-BoNT.¹²⁰ The clinician should evaluate each child and determine the risk of medical fragility and the benefit to be expected for each injection.¹²¹ However, recent literature and analysis reassures the safety for injecting BoNTs when used in all GMFCS levels.⁶²

Other Emerging Therapeutic Options

Therapies are necessary postinjection to maximize the outcome of the treatment. Splinting and casting orthotics can be enhanced and made more comfortable with the use of denervation by BoNT. Augmenting the uptake of BoNT has been done with electrical stimulation or with increasing the muscle activity postinjection.⁸⁹ A number of studies have investigated robotic postinjection, which holds promise.¹²² A treatment approach that balances the child’s and family’s goals with the resources and available modalities maximizes outcomes.

Figure 5. Improving motor function in CP: an integrative treatment-modalities graph. The graph (A) represents a schematic path of motor development in children with CP at each level of GMFCS. It represents in visual form the principles of common treatment options that can be considered in an interdisciplinary setting. This graph is not designed to represent a detailed or predictable clinical course, nor is it to be interpreted as a treatment algorithm. In the words of the developers: “The goal is to provide parents and caretakers, physicians and therapists with a means to plan treatments and interventions within the multidisciplinary treatment approach and to help answer questions concerning: What? When? How much? How long?” Guide to color codes: Dark green, basal curve representing all functional therapies. It forms the foundation to which all other therapies can be added on demand. Bright green, orthotics/adaptive aids. Yellow, oral medication. Orange, BoNT. Red, intrathecal baclofen. Blue, orthopedic surgery. (B) Expanded discussion of indications, principles, and limitations correspond to colors in the legend. Reprinted from Heinen et al. permission pending.⁶¹

Conclusions

CP is a common childhood condition that presents a spectrum of impairments and abilities that last into adulthood. There is reason to believe the incidence and prevalence are increasing. The lifelong consequences of the disorder represent and confer a great fiscal cost. The current management of CP is necessarily interdisciplinary. Treatment at present is symptomatic and aimed to decrease complications of the impairments and improve function. Spasticity is the most prevalent finding among affected children, and is variably susceptible to a range of manipulative, pharmacologic, surgical, and denervation strategies. BoNT A has been widely employed for management of spasticity, with experience confirming efficacy and generally good safety. Systematic reviews present a high level of evidence for tone reduction in both upper- and lower-extremity studies in children with CP.^{60,109,123-125} Practice parameters for children with CP reflect the evidence from the prominent and the frequent use of BoNT in children with CP.¹²⁶ Despite the accepted use and high level of evidence for the efficacy of BoNT, clinicians in the United Sates use the toxins without FDA approval. Further investigation into the optimal dosage of BoNT, procedural improvements to maximize precise administration to target muscles, and safety with long-term, repetitive use are warranted. In addition, head-to-head studies for the newly developed and marketed BoNTs can and will occur. Enhancement of the effect of toxins by combining technology and innovative therapy holds promise for improving functional and medical outcomes for children and adults with CP.

REFERENCES