Cerebral palsy: Treatment of spasticity, dystonia, and associated orthopedic issues

INTRODUCTION — Cerebral palsy (CP) refers to a heterogeneous group of conditions involving permanent motor dysfunction that affects muscle tone, posture, and/or movement. These conditions are due to abnormalities of the developing fetal or infant brain resulting from a variety of non-progressive causes. Although the disorder itself is not neurodegenerative, the clinical expression may change over time as the central nervous system matures. The motor impairment results in limitations in functional abilities and activity, which can vary in severity. Multiple additional symptoms often accompany the primary motor abnormalities, including altered sensation or perception, intellectual disability, communication and behavioral difficulties, seizures, and musculoskeletal complications [1].

The treatment of spasticity, dystonia, and associated orthopedic issues in patients with CP is discussed here. Other aspects of management and prognosis of CP are discussed separately. (See "Cerebral palsy: Overview of management and prognosis".)

The epidemiology, etiology, prevention, classification, clinical features, evaluation, and diagnosis of CP are also discussed separately:

●(See "Cerebral palsy: Epidemiology, etiology, and prevention".)

●(See "Cerebral palsy: Classification and clinical features".)

●(See "Cerebral palsy: Evaluation and diagnosis".)

FUNCTIONAL EVALUATION AND TREATMENT GOALS — Interventions for CP focus on maximizing the child's independence in daily functional activities and reducing the extent of disability. Assessment of the child's functional status guides treatment selection and allows for monitoring of change over time. Standardized tools for assessing motor function include:

●The Gross Motor Function Classification System (GMFCS)

●The Manual Ability Classification System (MACS)

●The Communication Function Classification System (CFCS)

Additional clinical tools include the Modified Ashworth Scale (table 1), which assesses tone, and the International Classification of Functioning, Disability and Health (ICF), which can be used to conceptualize the impact of CP on different elements of a child's health, function, and participation.

Evaluating the child's functional status and establishing treatment goals are discussed in greater detail separately. (See "Cerebral palsy: Overview of management and prognosis", section on 'Functional evaluation' and "Cerebral palsy: Classification and clinical features", section on 'Functional classification systems'.)

MANAGEMENT OF ALTERED MOTOR FUNCTION — By definition, all children with CP have altered motor function to some extent, affecting posture, coordination, and balance. Treatment interventions are typically offered from least to most invasive.

Early intervention — Therapeutic interventions should start at the time of the suspected diagnosis of CP to help maximize functional improvement related to neuromuscular plasticity of the developing brain. Referral for physical and occupational as well as other CP-specific therapies includes task-specific modalities and may be tailored to the age of the child [2].

Physical and occupational therapy — Physical therapy (PT) is an established part of treatment programs for CP and is an integral and vital part of management. Although its effectiveness in advancing physical function is uncertain, PT has an important role in promoting range of motion, positioning, stamina, and coordination, all of which can directly impact mobility and, with more severe forms of CP, transfers. Occupational therapy (OT) is equally important and focuses on fine motor skill development and improving capacity for self-care or caregiver support of tasks of daily living. As such, targeted activities include skills such as feeding, toileting, dressing, personal hygiene, and fine motor tasks, which are important for school success.

PT/OT are typically started early and continued by the parents or caregivers at home between sessions [2,3]. Session frequency, length, and duration are based on the child's needs. Services may be rendered at school (as part of an individualized educational plan), in the community, or in both settings. Intensive programs, which may include combinations of occupational, physical, and speech therapy, can be incorporated into the child's standard therapy schedule.

The efficacy and evidence base varies substantially among different types of therapeutic approaches. The best supported approaches are [2,4,5]:

●Bimanual training for hemiplegic CP – Trains the child in the use of two hands together through repetitive tasks [6-8].

●Constraint-induced movement therapy – For children with hemiplegic CP, constraint-induced movement therapy (CIMT) promotes improved motor and sensory function of the affected limb by encouraging its use through intermittent restraint of the unaffected limb during therapeutic tasks [6-10]. The method of restraint varies from holding a child's hand to casting, and the length of time in the restraint varies from 1 to 24 hours a day. "Forced use" is a variation of CIMT in which the limb's use is encouraged only by placement of the contralateral restraint; no additional therapeutic tasks are assigned to the affected limb [11-13]. Because the methods and outcomes used varied considerably among these trials, it is unclear which specific CIMT techniques are most useful [14,15].

●Context-focused therapy – Promotes successful task performance by changing the task or environment rather than changing the child's approach. A single randomized trial showed that this was as effective as child-focused therapy, in which there is a focus on remediation of motor impairment through practice of functional activities [16].

●Goal-directed/functional training – Focuses on activities based on goals set by the child, using a motor-learning or task-specific approach [8,17-20].

●Occupational therapy after botulinum neurotoxin (BoNT) treatment – Uses a variety of therapeutic approaches delivered in conjunction with BoNT treatment [21-24]. (See 'Botulinum toxin' below.)

●Aerobic exercise – Fitness training incorporating aerobic conditioning may improve both aerobic capacity and gross motor function [25-27]. Exercise programs are only effective in individuals who have sufficient motor skills to permit participation and are beneficial only as long as training continues.

●Other – Other approaches supported by at least moderate quality evidence are home programs for improving motor activity performance or self-care [17]. Equine-assisted therapies (eg, hippotherapy) may improve ambulation and gross motor function [28].

Although many of these approaches have been found to be effective in randomized trials and systematic reviews, the quality of the evidence is low to moderate for most, and clinical decisions must be based on individual patient characteristics and available resources [4,29].

In a comprehensive review, therapeutic approaches that were not recommended because the available evidence suggests they are ineffective include passive neurodevelopmental therapy (NDT), sensory integration therapy (SIT), craniosacral therapy, and hip bracing [2,29,30]. NDT is the direct passive handling of limbs by the therapist, with guidance intended to normalize the patterns of movement. It is not recommended because of the availability of more effective alternatives [2,29-32]. Any gains in range of motion that are achieved during an NDT session do not carry over. Better functional gains are achieved with active motor-learning approaches [30,33], and BoNT is more effective than NDT for managing spasticity [29]. Similarly, systematic reviews conclude that SIT is not effective and should be replaced by effective evidence-based alternatives such as those listed above [2,29,34].

Many other interventions are supported only by very low-quality or inconclusive evidence, precluding a recommendation for or against their use [29]. This includes animal-assisted therapy [35], conductive education [36], taping [37], Feldenkrais therapy [38], hydrotherapy, massage, oral-motor therapy, seating and positioning, stretching, TheraSuits, and whole-body vibration [29].

Orthotics and other devices — Numerous devices are available to help promote function, mobility, and participation for children with CP. These include:

●Braces/orthotics – Many children with CP benefit from bracing to enhance musculoskeletal alignment, promote function, and delay or prevent joint contractures [39,40]. Ankle-foot orthotics are the most frequently used devices, followed by orthotics for the wrist and hand. Orthotics for the knees or hips can also be useful to assist with ambulation or balance. Some braces are intended to be worn with specific activities, such as walking, using a gait trainer, or weight bearing in a stander, while others are intended to be used in sleep to promote joint alignment. Devices that use functional electrical stimulation via foot drop stimulation (eg, Bioness, Walkaide) have been shown to improve gait performance in adult patients with hemiplegia secondary to stroke, and they may be helpful to children with hemiplegic CP [41].

●Standers – As a child grows, functional equipment is often introduced to allow for age appropriate participation and to promote health. For children who are unable to weight bear independently, standers allow for participation in upright activities. At the same time, they promote bone health and can have positive secondary impacts on gastrointestinal motility and respiratory health.

●Seating systems – A variety of seating systems may be used to optimize positioning during activities, from eating meals at home to participating in the classroom. Seating systems may allow for better head positioning, enhance upright posture, or tilt to allow for interventions or rest.

●Mobility devices – For the child who is unable to ambulate independently or for long distances, mobility devices are critical. Wheelchairs can be introduced as early as two or three years. Most children who need a wheelchair start with a manual chair that allows another person to push them. Over time, if the child has the cognitive, visual, and motor abilities to steer a chair independently, an electric version may be appropriate and provide the child with more autonomy. There are numerous adaptations that allow a child who might have motor difficulty manipulating a joystick to steer, including knee and head controls. Lightweight adaptive strollers are less cumbersome than wheelchairs and are useful for quick transportation in the community and easier to accommodate in most vehicles.

Many specialty centers offer specialized clinics for evaluation of equipment and/or augmentative communication needs. These can be invaluable resources to the primary care clinician in identifying the most appropriate piece of equipment for the child and can also assist the caregivers in navigating the complex insurance regulations that can impact the timing and range of equipment available.

TONE MANAGEMENT — Pharmacologic and/or surgical interventions may help to reduce the altered tone associated with CP and improve functioning. Characterizing the nature and severity of the altered tone is necessary to help guide treatment options. Spasticity and dystonia may be more amenable to treatment than weakness or dyscoordination. (See "Cerebral palsy: Overview of management and prognosis", section on 'Assessment of tone'.)

When deciding whether to initiate or escalate treatments for altered tone, understanding goals of the child and caregiver are important. Functional considerations include:

●Is the altered tone causing considerable functional impairment?

●Is the altered tone causing chronic pain or discomfort?

●Does the altered tone make caregiving challenging (eg, positioning, hygiene, or dressing)?

●Is the child's sleep disrupted as a consequence of their altered tone?

●What is the level of concern of the patient and parent or caregiver?

Spasticity

General approach — There are many options available for the treatment of spasticity for children with CP. Treatments options may vary according to several factors including the child's age, goals for treatment, comorbid conditions, risk of adverse effects from treatment, regional distribution of spasticity, and the response to any prior treatments.

Physical and/or occupational therapy (PT/OT), position optimization, and orthotic use should be integrated into care for all patients (see 'Physical and occupational therapy' above and 'Orthotics and other devices' above). The general approach to additional pharmacologic and surgical treatments is as follows:

●For children with generalized spasticity, oral antispasticity drugs are usually the first-line treatment (eg, baclofen or benzodiazepines). (See 'Oral antispasticity drugs' below.)

●For treatment of localized or segmental spasticity, botulinum neurotoxin (BoNT) injections are appropriate. Alcohol block can be used in conjunction with BoNT or as an alternative for patients who do not tolerate or respond to BoNT, but there is less evidence to support its use. (See 'Botulinum toxin' below and 'Alcohol block' below.)

●For patients with spastic diplegia who have mild to moderate motor impairment (Gross Motor Function Classification System [GMFCS] level II or III), no significant weakness, and are cognitively able and motivated to participate in post-surgery rehabilitation, selective dorsal rhizotomy (SDR) may be performed as an alternative to or along with antispasticity therapy to improve gait. (See 'Selective dorsal rhizotomy' below.)

●For severely affected children (GMFCS level IV or V) who have significant side effects with oral antispasticity drugs or who do not achieve adequate response despite maximal dose, an intrathecal baclofen pump may achieve better control of spasticity. SDR is an alternative option for nonambulatory children to address refractory spasticity, provide comfort, and promote participation in activities of daily life [42]. (See 'Intrathecal baclofen' below and 'Selective dorsal rhizotomy' below.)

Oral antispasticity drugs — Baclofen and benzodiazepines have been most extensively used and are considered first-line agents to treat spasticity in children with CP [43,44]. Second-line agents include dantrolene and tizanidine.

●Baclofen – Baclofen, a gamma-aminobutyric acid agonist, is widely used for treatment of spasticity. Small randomized trials in children with CP demonstrated that oral baclofen improves spasticity and range of motion and may modestly improve function [29,43,45,46]. Baclofen may be preferred over benzodiazepines due to better tolerability and lower risk of abuse. However, high-quality evidence supporting its effectiveness is lacking.

Weight-based dosing of baclofen is started at a dose of 0.5 mg/kg per day given enterally in two or three divided doses. The total daily dose can then be titrated up if necessary in weekly increments of 0.5 mg/kg per day as tolerated to a maximal dose of 2 mg/kg per day. Titration stops once an effective dose is reached; not all children require the maximal dose. The rate of titration can be slowed if side effects develop. Doses higher than 2 mg/kg per day may be necessary in some patients, depending on the response and tolerance [47].

In older children and adolescents, initial baclofen dosing starts at 5 mg two or three times per day, increased at weekly intervals to a typical maximum of 20 mg three times per day. However higher amounts, above 60 mg per day, can be administered with close monitoring.

Common side effects include sedation, constipation, urinary retention, and nausea. Other reported side effects include confusion, hypotonia, ataxia, and paresthesias. Side effects may be minimized by slow titration but ultimately may limit treatment. Seizures and hallucination may occur if the drug is discontinued abruptly [48].

●Benzodiazepines – Benzodiazepines (eg, diazepam) are widely used to treat spasticity. Small, randomized trials in children with spastic CP have demonstrated that diazepam reduces hypertonia, improves passive range of movement, and increases spontaneous movement in the short term [29,43,49].

Diazepam is given enterally at a total daily dose of 0.01 to 0.3 mg/kg divided in two to four doses (maximum 10 mg per dose). The dose is titrated gradually to effect and tolerability. Side effects include sedation, weakness, ataxia, memory disturbances, and dependence after prolonged use.

If diazepam is not effective or well tolerated, other benzodiazepines can be trialed (eg, lorazepam, clonazepam). Individual patients may respond better to one benzodiazepine than another. Some patients with CP may require regularly scheduled benzodiazepine for ongoing spasticity management, whereas others may use the medication only as needed (eg, for spasms or sleep disruption).

●Dantrolene – Dantrolene is used less frequently than baclofen or benzodiazepines in children with CP. Treatment with dantrolene may decrease tone, tendon reflexes, and scissoring and may improve range of motion, but there is little evidence of objective functional improvement [29,50].

Side effects are common and include weakness, drowsiness, lethargy, paresthesias, nausea, and diarrhea [43]. The use of this drug is limited by hepatic toxicity. In a report of 1044 patients treated with dantrolene for at least 60 days, hepatic toxicity occurred in 1.8 percent [51]. When additional cases reported after marketing of the drug were included in the analysis (total of 50 cases), the case-fatality rate for dantrolene-associated liver disease was 28 percent. Liver disease resolved in the remainder after discontinuation of the drug. Prior to considering starting dantrolene, potentially hepatotoxic medications and supplements should be reviewed and weaned if possible. When dantrolene is used for treatment of spasticity in children, routine monitoring of hepatic function is necessary.

●Tizanidine – Tizanidine is an alpha-2 adrenergic agonist agent that decreases spasticity by increasing presynaptic inhibition. There is limited experience using tizanidine in children with CP, thus it is not a first-line agent. Based on the limited available evidence, tizanidine appears to reduce spasticity compared with placebo [29,52]. When used in combination with BoNT, side effects appear to be less than the combination of BoNT plus baclofen [53]. Long-term safety data are not available.

Reported side effects include dizziness, drowsiness, gastrointestinal upset, and vomiting. Liver injury ranges from hepatitis to acute liver failure. As with dantrolene, tizanidine should not be combined with other hepatotoxic medications and supplements.

Botulinum toxin — Injection of BoNT into affected muscles blocks the presynaptic release of acetylcholine from motor endplates of the lower motor neuron at the myoneural junction and decreases tone by limiting muscle contraction. It is injected into several areas of the muscle. Injections must be repeated every three to eight months to maintain the effect. The potency of the different BoNT preparations (abobotulinumtoxinA, incobotulinumtoxinA, onabotulinumtoxinA, rimabotulinumtoxinB) differs, and they should not be used interchangeably. Most reports suggest that these drugs are generally safe, but there have been case reports of life-threatening systemic toxicity. (See 'Complications' below.)

Patient selection — BoNT is used in patients who have increased muscle tone that interferes with function or is likely to lead to joint contracture with growth. BoNT is used for a variety of focal indications including spastic equinus deformity (treated by injections into the gastrocnemius-soleus muscles), knee and hip flexion spasticity, and spasticity of the upper extremity [43,54-58]. The treatment can be helpful for patients with a wide range of severity (GMFCS level I to V) [22,59].

Patients <4 years old and without fixed contractures are expected to have the most favorable response to BoNT treatment because the musculoskeletal and nervous systems are most pliable in growing children [60,61]. Nonetheless, BoNT also may be useful to treat focal spasticity in older children and adults.

For patients with diffuse hypertonia, BoNT can be injected into multiple muscles to treat the main foci of the generalized spasticity. Ideally, only two or three muscles will require treatment at one time. Fixed contractures are not amenable to BoNT therapy, although BoNT injections may facilitate serial casting. (See 'Casting' below.)

Treatment — Prior to treatment, a complete baseline assessment should be performed when possible by a multidisciplinary team to determine appropriateness for intervention and to establish specific treatment goals. The evaluation should include range of motion, strength, and assessment of selective motor control, muscle tone, and function.

Dosing of BoNT depends on the muscles injected, body weight, muscle bulk, the number of muscles being injected simultaneously, and the patient's response to previous therapy [61,62]. No reliable potency standard has been established for BoNT, and the biological effects may vary among preparations from different manufacturers [54]. As a result, the different BoNT preparations (abobotulinumtoxinA, incobotulinumtoxinA, onabotulinumtoxinA, and rimabotulinumtoxinB) should not be used interchangeably, and separate dosing schemes must be established for each.

The patient should be reevaluated six to eight weeks after treatment. The effect typically lasts three to eight months, after which treatment can be repeated.

Efficacy — BoNT temporarily reduces spasticity and delays the shortening of affected muscles, as shown by case series, open label studies, and clinical trials [21,43,59,63-69]. Clinical trial evidence suggests that BoNT in combination with PT/OT modestly improves some functional outcomes, although it does not appear to improve gait [21,29,43,68,70].

A 2010 systematic review identified 10 randomized trials evaluating the efficacy and safety of BoNT as an adjunct to treatment in children with upper limb spasticity due to CP [21]. Most trials evaluated the effects of BoNT combined with PT/OT compared with PT/OT alone. BoNT improved short-term measures of spasticity and range of motion. However, in pooled analysis of three trials (109 patients), there was little improvement in self-perception or caregiver perception of performance in activities of everyday living at three months (as assessed by the Canadian Occupational Performance Measure).

In the largest pediatric trial evaluating BoNT (which was not included in the systematic review), 235 children with spastic CP and dynamic equinus foot deformity were randomized to one of two doses of BoNT or placebo. Both doses of BoNT improved muscle tone as assessed by the Modified Ashworth Scale (MAS) (table 1) [68]. At four weeks following injection, the mean MAS score improved from 3.1 to 2.2 among patients treated with a BoNT dose of 15 units/kg/leg (mean difference -0.97; 95% CI -1.18 to -0.76), from 3.1 to 2.3 in patients treated with a dose of 10 units/kg/leg (mean difference -0.86; 95% CI -1.07 to -0.65), and from 3.2 to 2.6 in patients who received placebo (mean difference -0.48; 95% CI -0.69 to -0.27). Global clinical status, which was assessed using a 9-point scale from -4 (markedly worse) to +4 (markedly improved), was also improved (mean scores at week 4 were +1.54, +1.50, and +0.73 for the BoNT 15 units/kg/leg, BoNT 10 units/kg/leg, and placebo groups, respectively). In a secondary analysis of this trial, 77 percent of children treated with BoNT achieved their primary function goal during follow-up compared with 62 percent of those treated with placebo [71]. Goals were chosen by the parents/caregivers, with the most common goals being improved walking pattern, improved balance, and decreased frequency of falling.

Treatment benefits may be subtle and may not be obvious to the patient or caregivers. A randomized trial that included 33 children with CP and spastic diplegia showed no difference in the caregiver's satisfaction with the child's function, despite documentation of improvement in spasticity and function in the botulinum-treated group, using a variety of objective measures [72]. These observations underscore that clinicians should communicate realistic expectations prior to embarking on treatment.

The long-term effects of BoNT were evaluated in a randomized controlled trial in 64 children with spastic CP treated with BoNT or placebo for two years [73]. Changes in motor function were assessed with the Gross Motor Function Measure (GMFM) score and compared with pretreatment baseline in both groups. Changes in functional abilities were assessed using the pediatric evaluation of disability index (PEDI) score. There was a trend toward greater improvement in both GMFM and PEDI scores among placebo-treated children over the two-year study period (ie, motor function and functional ability appeared to worsen with long-term BoNT treatment); however, the findings were not statistically significant. The study was limited by its small sample size and high rates of withdrawal and loss to follow-up (30 and 9 percent of patients, respectively).

The effect of BoNT treatment on bone and soft tissue are not clear. Initial hopes that early and continued treatment with BoNT might modify the effect of spasticity on bone and soft tissue and reduce the need for later orthopedic surgery have not been born out in clinical trials. Based on the available evidence, long-term BoNT treatment does not appear to prevent contractures or reduce the need for orthopedic surgical intervention. In a randomized controlled trial in 64 children with spastic CP treated with BoNT or placebo for two years, the proportion of children who required orthopedic surgery during follow-up was similar in both groups (43 and 46 percent, respectively) [73]. Similarly, in a prospective study of long-term use of BoNT in 94 children with CP (median length of treatment 18 months), there was gradual deterioration in range of motion, despite beneficial effects on muscle tone after each injection [74]. The study did not include a control group, so it is unclear whether the range of motion deteriorated faster or slower than would have occurred without BoNT.

Complications — The majority of complications after BoNT injection are transient, mild, and self-limiting. Fever is common and may last for one to three days. Other common side effects include transient pain, local irritation, and bruising. A few patients experience temporary weakness and loss of function in the injected area, including urinary retention or constipation due to sphincter dysfunction for lower extremity injections. These symptoms usually last <2 weeks and may be related to a dose that is too high or, in the case of weakness, previously unappreciated underlying muscle weakness.

Severe life-threatening systemic adverse effects, including difficulty swallowing or breathing, have been reported as rare complications [75-77]. Clinicians should be alert for the potential of systemic effects, including dysphagia, dysphonia, weakness, or dyspnea, which may occur up to several weeks after treatment. All preparations of BoNT include a boxed warning regarding potential for systemic effects. Some of these complications appear to be caused by inadvertent overdosing because potency determinations expressed in "units" varied among BoNT products. In 2008, the US Food and Drug Administration mandated changes in drug names, designed to emphasize these differences in drug potency and prevent medication errors, and updated the drug safety information for health care professionals [78].

The types and frequency of BoNT complications were illustrated in an observational study of 591 children with CP undergoing 2219 injections at a single center [77]. Adverse events attributed to systemic spread of the drug occurred in 3.6 percent of injections. Severe systemic adverse events requiring hospitalization occurred in 0.7 percent of injections and included worsening dysphagia, generalized weakness, lower respiratory tract infection, choking, and aspiration. Risk factors for experiencing an adverse event following BoNT injection in this series included severe CP (GMFCS level IV or V), known underlying dysphagia, and history of aspiration pneumonia in the past.

Most clinical trials reported similar rates of adverse events in children treated with BoNT and placebo [76]; however, the available clinical trial data are limited by small sample size and limited duration of follow-up.

Secondary nonresponse — Some patients become less responsive to BoNT over time (a phenomenon termed "secondary nonresponse") [79-82]. In some cases, this is associated with the development of neutralizing antibodies, though neutralizing antibodies do not always cause treatment nonresponse. Neutralizing antibodies to the toxin develop in 3 to 10 percent of patients after repeated injections [83-85]. The relative immunogenicity of different BoNT type A and type B preparations has not been fully established. In addition, for patients with neutralizing antibodies to one preparation (eg, a type A BoNT preparation), it is unclear whether use of another preparation (eg, a type B preparation) will be effective.

Alcohol block — Alcohol blocks with either phenol or ethanol can be used in conjunction with BoNT injections since the combination may achieve better results and the effect may last longer than either treatment alone; however, the risk of adverse effects is also increased [86]. Alcohol blocks can also be used as an alternative for patients in whom BoNT is poorly tolerated or ineffective (eg, due to development of neutralizing antibodies). Phenol and ethanol are neurolytic agents that, when given as a motor point injection, provide temporary reduction in spasticity. Clinician experience and institutional protocol typically guide selection of agent. Phenol is used most frequently.

Alcohol blocks are used specifically for nerves with only motor function (hip adductor and hamstrings most commonly, though the biceps and brachioradialis are reported) to avoid the discomfort of neurolysis near sensory nerves. Potential complications with phenol blocks include necrosis of the skin and surrounding muscle. Accidental intravascular injection can result in tinnitus and flushing. Neuritis can occur after partial destruction of a somatic nerve with subsequent regeneration; the dysesthesia or hyperesthesia that results may be worse than the original pain.

Data on the efficacy and safety of phenol blocks in children with CP are limited [87-90]. The available data suggest that, compared with BoNT, phenol blocks are associated with less improvement in spasticity and greater risk of side effects. Thus, the modest evidence regarding efficacy must be considered in the context of safety risk.

Intrathecal baclofen — For severely affected children who have significant side effects with antispasticity drugs or who have persistent and severe spasticity despite maximal doses, intrathecal baclofen may be an option and is an effective intervention [5]. Intrathecal baclofen administered continuously via a pump achieves higher cerebrospinal fluid (CSF) drug levels as compared with oral administration. CSF drug levels are typically low with orally administered baclofen because of poor lipid solubility.

Intrathecal baclofen reduces spasticity and clonus in severely affected patients, but it is also associated with risk for substantial complications [91-95]. Therefore, this treatment is generally restricted to patients with severe spasticity that is not responsive to other measures. Guidelines for determining patient selection and the optimal dose and placement of the catheter tip have not been established [96].

In a case series of 35 children with severe CP treated with a continuous infusion of intrathecal baclofen for 18 months, patients experienced early improvement in pain and sleep and later improvement in social function and mobility [97]. However, treatment was stopped in three additional patients because of agitation or infection. Another study described the effects of intrathecal baclofen for 18 months in children with severe CP, using a group awaiting treatment as controls [98]. Children treated with intrathecal baclofen had reduced tone and spasms and improved perceived comfort and ease of nursing care. The main side effect was increased constipation.

Side effects of intrathecal baclofen include lethargy, confusion, and hypotonia, which appear to be dose related and occur in up to 50 percent of patients [99,100]. Catheter-related complications include infection, seroma, and CSF leak [101,102]. Pump malfunction can also occur and must be considered promptly as serious life-threatening symptoms may arise with baclofen overdose or withdrawal.

Selective dorsal rhizotomy — SDR is a surgical procedure that selectively divides parts of the dorsal lumbosacral roots of the spinal cord [103]. This interrupts the afferent limb of the reflex arc on the sensory side, thus reducing spasticity without causing paralysis. This technique may provide a small gain in function.

Which children are most likely to benefit from SDR is uncertain. The available evidence to guide treatment is limited, and practice varies considerably. There are two general categories of patients who may benefit from SDR:

●Children with relatively high levels of physical function – This includes patients who are functionally limited by spasticity but with some degree of independent ambulation and with potential to make functional gains following surgery (eg, patients with spastic diplegia) [104-107]. Patient selection should be rigorous since not all patients will benefit. Historically, the optimal patients for SDR were thought to be children with spastic diplegia resulting from periventricular leukomalacia; however, over time, other etiologies have also been included (eg, hereditary spastic paraplegias). The patient's age is another important consideration. Young children (age >2 years to <10 years) are generally considered to be optimal candidates since they are young enough to relearn appropriate motor patterns for ambulation. Adequate cognition is necessary to allow for active engagement of the child with the intensive rehabilitation required postoperatively. Scrutiny of the degree of underlying muscle weakness is essential since long-term functional motor disability can be worsened if SDR is pursued in children with underlying muscle weakness.

The efficacy of SDR in this setting is supported by observational data and a few small clinical trials [104-114]. In a meta-analysis of three randomized controlled trials comparing SDR plus PT with PT alone in a total of 90 children with spastic diplegia who were primarily ambulatory (most were <8 years old and most had a GMFCS level of II or III), spasticity outcomes were assessed with the Ashworth scale (which ranges from 0 to 4, with higher scores reflecting more severe spasticity (table 1)) and motor function was assessed with the GMFM-66 score (which ranges from 0 to 100, with higher scores representing greater motor function) [104]. At follow-up of 9 to 12 months, children who underwent SDR plus PT had lower Ashworth scores and slightly higher GMFM-66 scores compared with those who received PT alone (mean differences of -1.2 points for Ashworth score and 2.6 points for GMFM-66 score; 95% CIs were not reported, but both findings were statistically significant with a p value of <0.01) [104]. Thus, SDR had a clinically apparent effect on spasticity but only a modest effect on motor function. The improvement in spasticity correlated with the proportion of dorsal root tissue that was transected. No serious adverse events were reported.

In a subsequent study, 137 children (ages 3 to 9 years) with spastic diplegia and GMFCS level II to III who underwent SDR at one of five centers in England were followed prospectively for two years [108]. GMFM-66 scores improved from presurgery baseline slightly more than would be expected based on the trajectory of normative data in children with CP who did not undergo SDR (mean improvement was 3.2 per year in the study population compared with an expected rate of improvement of 1.9 per year). Greater improvements were observed in children with GMFCS level II compared with GMFCS level III. Thus, though the surgery appeared to have a measurable benefit on motor function, the effect was small. The study also found small improvements in some but not all measures of quality of life. Approximately 10 percent of children in the study experienced at least one adverse event, most commonly wound infection and persisting dysesthesias in the feet and legs. None of the adverse effects were severe.

Long-term outcomes of SDR are described in several observational case series, which included ambulatory and nonambulatory patients (GMFCS levels I through III) [109-115]. In the largest of these series, durable improvements in lower-limb muscle tone, gross motor function, and performance of activities of daily living were found in each GMFCS group (levels I through III), although individual results varied and nearly 40 percent needed additional treatment with BoNT injection or lower extremity orthopedic surgery [110]. Patients with spastic diplegia tended to have better outcomes compared with children with quadriplegia.

●Severely affected nonambulatory patients – The use of SDR in the setting of severe motor impairment is emerging. Patients with severe spasticity and contractures that cause significant discomfort and/or interfere with sitting and general caretaking (eg, bathing, positioning, toileting) may have some benefit from SDR, and some centers offer surgery as an option for these patients. Patients with severe spasticity and contractures that cause significant discomfort and/or interfere with sitting and general caretaking (eg, bathing, positioning, toileting) may have some benefit from SDR, and some centers offer surgery as an option for these patients [116-118]. Many patients in this category have other comorbidities (eg, intellectual disability, seizure disorder). The goals of surgery in this setting are to improve comfort, improve sitting stability, and to ease the challenges of caretaking. SDR in severely affected patients generally requires greater extent of nerve root division, and some patients may experience troublesome weakness as a result.

Observational studies in severely affected children (GMFCS level IV or V) suggest that SDR in combination with PT is associated with improvements in spasticity, range of motion, and urinary continence [116-118]. In a study comparing outcomes in a consecutive series of 71 children who underwent SDR with a group of 71 children (matched by age and preoperative GMFCS level) who underwent intrathecal baclofen pump placement, both treatments decreased tone and improved range of motion [118]. SDR was associated with greater improvements in tone and gross motor function. Additional studies are needed to better understand the risks and benefits of SDR in this patient population.

Dystonia

Pharmacologic treatment — Many oral agents have been used to treat dystonia, but they have not been extensively studied in rigorous controlled trials (table 2). The therapeutic window for most of the oral agents is narrow, efficacy is often modest, and side effects frequently limit clinical benefit [5,119,120]. Parenteral agents are more difficult to administer than oral agents.

Commonly used agents include anticholinergics (eg, trihexyphenidyl), benzodiazepines (eg, clonazepam), oral baclofen, and botulinum toxin injections. Intrathecal baclofen is typically reserved for those who are unable to tolerate or are unresponsive to oral pharmacologic agents.

The approach to treatment of dystonia is reviewed in detail separately. (See "Treatment of dystonia in children and adults".)

Deep brain stimulation — Deep brain stimulation (DBS) is generally reserved for patients with severe dystonia that has not responded to medical therapy. DBS is achieved through a microelectrode that stimulates the internal segment of the globus pallidus. It has been used for adults with severe primary dystonia (not CP) who fail treatment with pharmacologic agents and BoNT injections. Meta-analyses of DBS in patients with CP suggest that it may be effective, but further study is required to identify optimal patients for this option [120-123]. DBS for dystonia is discussed in a separate topic review. (See "Treatment of dystonia in children and adults", section on 'Surgical therapy'.)

ORTHOPEDIC ASSESSMENT AND INTERVENTIONS — Orthopedic interventions are directed at prevention or correction of joint deformities and maximizing function. Treatment selection is based on clinical evaluation, severity of symptoms, the efficacy of the intervention, and patient and caregiver goals. The goals of orthopedic intervention are to improve positioning and gait and to prevent pain.

Gait analysis — Gait analysis, using physical examination, videotaping, force plates, electromyography, and computerized analysis of limb motion, can be used as an assessment tool to identify abnormalities in muscle function and limb alignment and to evaluate both the indications for and the effects of surgery [124-126].

Casting — Serial casting in the upper and lower extremities is used to stretch the shortened muscles, correct joint deformities, and improve range of motion. Systematic reviews of observational studies and a few small randomized trials reveal that ankle casting results in small improvements in range of motion and short-term improvements in contractures, but it is unclear if these effects are clinically meaningful over the long term [5,29,127-129]. A small clinical trial showed that the addition of wrist casting to botulinum neurotoxin (BoNT) injections and occupational therapy (OT) resulted in greater improvement of muscle tone, spasticity, and passive range of motion at the 12-week follow-up [130]. Longer-term benefits are uncertain.

Active goal-directed and strength training following casting are used to help sustain improvements in function [20,129].

Surgery for contractures and bony deformities — Children with CP can develop fixed joint deformities and contractures that lead to alterations in the normal growth and development of the skeletal system. Surgical procedures for children with CP can be broadly divided into soft tissue procedures on tendons and muscles and bone procedures.

Soft tissue surgery — Lengthening, release, or transfer of muscle tendon units can improve function and delay or prevent deformities. These procedures are typically considered when abnormal muscle action or contractures interfere with function or care. They are often performed as part of a single-event multilevel surgery (SEMLS). (See 'Multilevel surgery' below.)

Muscle-tendon surgeries result in small improvements in gait function. As an example, one study prospectively evaluated the effect of muscle-tendon surgery in 30 patients with spastic diplegia [131]. Operations were performed at a mean age of 8.7 years. Gait velocity and stride length increased by 25 and 18 percent over preoperative values, respectively, nine months after surgery and were maintained for two years. However, the Gross Motor Function Measure (GMFM) score changed little after surgery.

In a nonrandomized prospective study comparing upper extremity tendon transfer surgery (n = 16) with BoNT injections (n = 13) or regular physical therapy (n = 10) in 39 children (ages 4 to 16 years), upper extremity tendon transfer surgery resulted in a greater degree of functional mobility and improved movement-related quality of life scores compared with nonsurgical management [132].

Bone surgery — Normal growth and development of the skeletal system depends on the bones being exposed to normal forces. When these forces are altered by abnormal muscle tone and abnormal weight-bearing patterns, secondary skeletal deformities can develop. Muscle-tendon surgery alone may not be sufficient to correct or prevent these deformities. Bone procedures may be used to permanently eliminate joint motion and correct deformity by slowing changes to the growth plate (eg, hemiepiphysiodesis) or by changing alignment and/or fusing the joint (eg, osteotomy).

Multilevel surgery — SEMLS is widely used in the management of children with CP. It involves multiple corrections of soft tissue and/or bony abnormalities at different anatomic levels performed in one surgery under one anesthesia [133-136]. Historically, serial surgeries were performed over a number of years. The advantage of SEMLS is that the child requires fewer surgeries and a shorter recovery period.

Hip disorders — Disorders of the hip are common complications of CP. These include progressive hip dysplasia, subluxation, and dislocation with degeneration and pain [137]. Although the hip is usually normal at birth, progressive dysplasia results from abnormal growth of the proximal femur and acetabulum. Nonambulatory children with quadriplegia are at the highest risk of hip disorders. Nearly half of nonambulatory children with spastic quadriplegia develop hip subluxation by five years of age as compared with 21 percent of those with diplegia [138].

Screening and management of hip disorders in children with CP is outlined in a Care Pathway of the Academy of Cerebral Palsy and Developmental Medicine [139]:

●Prevention – The most effective preventive strategy is early detection and appropriate management. Small observational studies have found decreased rates of early hip subluxation in children who start an abducted standing program before the age of 12 months [140,141].

Interventions to control spasticity do not appear to have an effect on hip subluxation [142]. A 2017 systematic review of available trials and observational studies identified no intervention with a substantial treatment effect at prevention hip subluxation [143]. In one clinical trial included in the systematic review, 90 children at risk for hip dysplasia were assigned to combination treatment with BoNT and hip bracing or to no treatment [144]. Similar rates of progressive hip displacement were seen in the treatment group as compared with controls.

●Screening – Because earlier surgical intervention is associated with better outcomes, we suggest that all children with CP undergo routine surveillance for hip dysplasia, consisting of physical examination and radiologic screening [29,138,145-147]. Screening programs appear to reduce the rate of hip dislocation when paired with appropriate and timely interventions [148]. Radiologic screening is performed with a supine anteroposterior radiograph of the pelvis with a bolster placed under the knees. Physical examination can occasionally detect clinical signs of subluxation in radiographically mild subluxation. The affected leg may be shortened and internally rotated, and there may be a palpable fullness in the gluteal region. The interval of screening varies depending on the child's motor function and other risk factors [149]. In nonambulatory patients, radiographs are generally obtained every 6 to 12 months.

●Index of suspicion – In addition to routine surveillance, the clinician should maintain a high index of suspicion for hip dislocation/subluxation in children and adults with CP who present with acute or chronic pain of uncertain etiology. Radiologic evaluation of the entire leg should be performed if hip pain can be elicited on physical examination, if passive range of motion is reduced compared with previous examinations, or for significant caregiver concern.

Leg pain in a child or adult with a dislocated or severely subluxated hip may also be due to a fracture [150]. (See "Cerebral palsy: Overview of management and prognosis", section on 'Bone health'.)

●Management – Treatment options for hip subluxation and dislocation include noninvasive abduction bracing, soft-tissue releases, reconstructive femoral and/or pelvic osteotomies, and salvage procedures such as valgus osteotomy of the proximal femur or proximal femoral resection. Bracing may help improve quality of life and delay need for surgery for some children [151]. In most children with hip subluxation, femoral and possibly pelvic osteotomies are indicated [152,153]. However, the surgical management required to prevent recurrent subluxation depends on the preoperative severity of the hip subluxation and acetabular dysplasia [154]. Routine hip surveillance and timely intervention are key to avoiding the need for salvage surgery [155].

Children who undergo soft tissue procedures alone are likely to need subsequent surgery, particularly if they have either severe spasticity (ie, Gross Motor Function Classification System [GMFCS] level IV or V) or radiologic evidence of hip dysplasia prior to intervention [156].

Neuromuscular scoliosis — Scoliosis is common in patients with CP, affecting approximately one-quarter to one-half of patients [157,158]. The incidence varies depending on the severity of motor impairment. Most children with severe motor impairment (GMFCS level IV or V) develop some degree of scoliosis by adolescence; in some patients, the onset may be as early as age 5 or 6 years [159]. In a study of 666 children with CP, the rate of moderate to severe scoliosis correlated directly with GMFCS level: 2 percent for GMFCS level I, 7 percent for GMFCS level II/III, and 30 percent for GMFCS level IV/V [157].

Neuromuscular scoliosis in patients with CP is generally progressive. It can result in pelvic obliquity and seating difficulties. Caregiving can become more challenging, and pain can be a complicating factor. In addition, scoliosis can have adverse long-term effects on cardiopulmonary and gastrointestinal function [160]. (See "Chest wall diseases and restrictive physiology", section on 'Kyphosis and scoliosis'.)

Unlike idiopathic scoliosis, the classic curve pattern in CP is a long C-shaped kyphoscoliotic curve (image 1). However, a wide range of spinal deformities can be seen.

Management of neuromuscular scoliosis is complex and best accomplished with a multidisciplinary team approach.

●Nonsurgical management – Nonsurgical management includes spinal bracing and use of molded wheelchair inserts to improve sitting balance and optimize comfort and positioning. Spinal bracing may improve sitting function [161,162]. However, progression is likely, particularly in patients with early onset (before age 10 years) and/or >40 degree curvature [158,159]. Antispasticity medications do not play a role in limiting progression of the curve but may help with overall comfort if muscle spasm is contributing to pain. Many children with significant neuromuscular scoliosis from CP are already taking medication for tone management. (See 'Spasticity' above.)

●Surgical management – Surgery for children with CP and scoliosis varies by age. Children who have a considerable amount of growth remaining may undergo growth-friendly spinal instrumentation (eg, growing rods) to prevent progression of severe curves [163-165]. For children closer to skeletal maturity, spinal fusion surgery is an option. Criteria for patient selection are not standardized. Reasonable candidates may include those who meet the following criteria [158]:

•>40 to 50 degree curvature that is progressive and interferes with sitting

•Age >10 years (though surgery can be considered at a younger age if the curvature is severe)

•Adequate hip range of motion to allow proper seating postoperatively

•Stable nutritional and medical status (though there may be considerable medical complexity)

The decision to pursue spinal fusion is multifaceted and should be grounded in the patient and caregiver goals and realistic expectations, as well as an understanding of the risks and benefits [166]. Because of the increased risks of spinal surgery in this population (as discussed below), surgery should be considered only when the expected benefits outweigh both the risks associated with surgery and the risks of not intervening (ie, the negative consequences of the natural history of progressive scoliosis). Preoperative evaluation should include a comprehensive medical evaluation, including assessment of cardiopulmonary function, nutritional status, and neurologic status. A multidisciplinary approach to surgery to anticipate potential risks and complications is desirable. This includes input from orthopedic surgery, general medicine, anesthesia, and other specialties as indicated [167].

Spinal fusion surgery for patients with CP generally involves multilevel internal fixation. The goal of surgery is to achieve solid bony fusion of the spine that corrects the curvature and results in a well-balanced spine with a level pelvis (image 1) [158]. It may involve posterior fusion only or anterior release with posterior fusion; surgery may be performed as a staged or same-day procedure [168]. Different surgical techniques have been described, and the technical details are beyond the scope of this topic review.

●Efficacy of scoliosis surgery – Data evaluating the efficacy of scoliosis surgery in patients with CP are limited to small retrospective cohort studies and case series [169-171]. In a systematic review of 51 observational reports, most studies reported good achievement of correction of the scoliosis deformity (typically measured by improvement in the Cobb angle (image 2)) [169]. There was a trend toward improvement in measures of caregiver satisfaction and quality of life; however, there was minimal to no improvement in gross motor or overall function. In a representative multicenter study that reported outcomes in 69 consecutive children and adolescents (mean age 13.5 years) with severe CP (GMFCS level IV or V) who underwent scoliosis surgery and were followed for five years, the Cobb angle improved from a mean of 82 degrees preoperatively to 28 degrees at one year postoperatively and then increased slightly to 32 degrees by five years [171]. Pelvic obliquity improved from a mean of 27 degrees preoperatively to 8 degrees at five years. There were modest improvements in some measures of caregiver-reported disease-specific quality of life, most notably in domains of personal care (mean improvement 7 points on a 100-point scale), positioning (mean improvement 9 points on a 100-point scale), and comfort (mean improvement 9 points on a 100-point scale) [172,173]. Data on long-term health outcomes, such as respiratory function, are limited. However, scoliosis surgery may reduce mortality rate, likely driven by reduction in subsequent respiratory complications. In a population-based registry of children with CP and scoliosis, mortality rates were higher in children treated nonsurgically than in those treated surgically (8.7 versus 5.3 per 1000 follow-up years) [174]. Respiratory cause of death during follow-up was more common in those treated nonsurgically (76 versus 38 percent).

●Complications – Reported complication rates of scoliosis surgery in patients with CP range from 25 to 50 percent [169,171,175]. These rates are considerably higher than in adolescents with idiopathic scoliosis; children with CP also tend to require longer duration of hospitalization postoperatively.

The most common complications include [169,175]:

•Surgical site infection

•Postoperative pneumonia

•Major blood loss

•Pseudoarthrosis (failure of fusion)

•Other hardware complications (eg, malposition, prominence, loosening of instrumentation)

Other reported complications include [169,171,175]:

•Neurologic injury

•Prolonged mechanical ventilation

•Pleural effusion

•Pneumothorax

•Pancreatitis

•Prolonged ileus

•Decubitus ulcer

Perioperative mortality is uncommon for children with CP undergoing scoliosis surgery, with reported rates ranging from 0 to 10 percent [169]. In a multicenter study of 808 patients undergoing spinal surgery for neuromuscular scoliosis, the mortality rate was 0.01 percent at 120 days and 1.1 percent by two years [176].

Risk factors for experiencing a major perioperative complication include [175,177]:

•Nonambulatory status

•Preexisting pulmonary compromise

•High degree of medical complexity

•Large intraoperative blood loss

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cerebral palsy" and "Society guideline links: Dystonia in children and adults" and "Society guideline links: Developmental screening".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or email these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient education" and the keyword[s] of interest.)

●Basics topics (see "Patient education: Cerebral palsy (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Initial evaluation and approach to treatment – All children with cerebral palsy (CP) have altered motor function, affecting posture, coordination, and balance. Treatment interventions should begin at the time of suspected diagnosis and focus on maximizing the child's independence in daily functional activities and reducing the extent of disability. Assessment of the child's functional status guides treatment selection and allows for monitoring of change over time. (See "Cerebral palsy: Overview of management and prognosis", section on 'Functional evaluation' and "Cerebral palsy: Classification and clinical features", section on 'Functional classification systems'.)

●Physical and occupational therapy – Physical and occupational therapy (PT/OT) are established and vital parts of treatment programs for CP. PT and OT are typically incorporated into care of all patients at a young age. (See 'Physical and occupational therapy' above.)

In addition, some patients with CP require devices to help promote function, mobility, and participation. These include braces, orthotics, standers, seating systems, and mobility devices. (See 'Orthotics and other devices' above.)

●Spasticity treatment options – Treatment options for spasticity may vary according to several factors, including the child's age, goals for treatment, comorbid conditions, risk of adverse effects from treatment, regional distribution of spasticity, and the response to any prior treatments. Our approach to managing spasticity in children with CP is as follows:

•Oral antispasticity medications – For children with generalized spasticity, we suggest oral antispasticity medications with baclofen or benzodiazepines as first-line therapy (Grade 2C). Dantrolene and tizanidine are alternative options. (See 'Oral antispasticity drugs' above.)

•Botulinum toxin – For treatment of localized or segmental spasticity, we suggest botulinum neurotoxin (BoNT) injections (Grade 2B). Dosing varies by the specific BoNT preparation used (abobotulinumtoxinA, incobotulinumtoxinA, onabotulinumtoxinA, and rimabotulinumtoxinB); they should not be used interchangeably. (See 'Botulinum toxin' above.)

•Alcohol blocks – Alcohol blocks can be used in conjunction with BoNT injections or as an alternative for patients who do not tolerate or respond to BoNT. (See 'Alcohol block' above.)

•Other options

-Selective dorsal rhizotomy – For patients with spastic diplegia who have mild to moderate motor impairment, no significant weakness, and are cognitively able to participate in post-surgery rehabilitation, selective dorsal rhizotomy (SDR) may be an option as an alternative or adjunct intervention to improve function. SDR may also be used for selected patients with severe motor impairment. (See 'Selective dorsal rhizotomy' above.)

-Intrathecal baclofen – For severely affected children who have significant side effects with oral antispasticity drugs or who do not achieve adequate response despite maximal dose, an intrathecal baclofen pump may provide better control of spasticity as well as improved comfort and positioning. (See 'Intrathecal baclofen' above.)

●Dystonia treatment options – Many oral agents have been used to treat dystonia, but they have not been extensively studied in rigorous controlled trials (table 2). The approach to treating dystonia is reviewed in detail separately. (See "Treatment of dystonia in children and adults".)

●Orthopedic interventions – Orthopedic interventions are directed at prevention or correction of deformities and maximizing function. Options include bracing, casting, and soft-tissue or bone surgeries. (See 'Orthopedic assessment and interventions' above.)

ACKNOWLEDGMENTS — The UpToDate editorial staff acknowledges Geoffrey Miller, MD and Laurie Glader, MD, who contributed to earlier versions of this topic review.