Hand Surgery
1st Edition

Reconstruction of the Spastic Hand
Ann E. Van Heest
Definition and Classification
Spasticity in the hand is a secondary peripheral manifestation of primary central nervous system (CNS) dysfunction. Classification of the spasticity depends on the type of primary CNS lesion, most commonly cerebral palsy, stroke, and traumatic brain injury. The peripheral effects of cerebral palsy, stroke, and traumatic brain injury are classified based on the topographic involvement and the type of tone that is present (1). Topographic involvement means the number of limbs that are involved: monoplegia (one limb), hemiplegia (one arm, one leg), diplegia (both legs), quadriplegia (both arms and both legs), and total body involvement (quadriplegia with mental retardation). The type of muscle tone disorder that is clinically manifested further classifies the disorder. Manifestations of CNS dysfunction can include spasticity, dystonia (athetosis), flaccidity, or mixed dysfunction. The amount of CNS control of the peripheral muscles is variable and is important prognostically; isolated or combined spasticity, dystonia, and flaccidity can be present.
Incidence and Natural History
Cerebral palsy is a nonprogressive CNS dysfunction that occurs in the perinatal period. Cerebral palsy is associated most commonly with prematurity and low birth weight (2) but can be due to neonatal asphyxia, intrauterine stroke, CNS malformation, infections, such as neonatal meningitis, kernicterus (Rh incompatibility), or postnatal brain injury that is due to closed head trauma or anoxic events.
Stroke is most common in individuals owing to a cerebral vascular accident that affects the middle cerebral artery, injures the motor and sensory strip in the brain, and leads to a hemiplegic pattern of deformity. Strokes affect 1 in 1,000 individuals per year, causing 200,000 deaths per year, and are a major cause of hemiplegia in adults in the United States. If the patient survives the initial 6 months after stroke, the average subsequent survival is more than 6 years. Within 6 months after stroke, recovery is usually stable for surgical assessment (3,4,5 and 6).
Traumatic brain injury affects 1 to 2 per 1,000 individuals per year in the United States, with most affected individuals younger than 40 years of age, and occurring secondary to motor vehicle accidents. Eleven percent of individuals with traumatic brain injury die. Traumatic brain injury is assessed by Glasgow Coma Scale, with younger patients with higher Glasgow Coma Scale scores having the best chances for survival. Patients who emerge from coma within 2 weeks have the best chance for neurologic recovery. Improvement in motor function improves at as long as 18 months after injury, with cognitive function continuing beyond 18 months (6).
In cerebral palsy, cerebral vascular accidents, and traumatic brain injury, a pattern of spastic hemiplegia is the most common manifestation of the CNS deficiency. This chapter focuses on the most common form of spasticity in the hand: spastic hemiplegia due to cerebral palsy. In general, the principles that are used to treat this disorder can be applied to other disorders that cause similar spasticity deformities.
Spasticity is characterized by a velocity-dependent increase in muscle tone. In upper motor neuron lesions, such as cerebral palsy, traumatic brain injury, and stroke, the normal inhibitory function is lost. Many neural pathways control stretch reflex excitability, and a malfunction in any of them could produce spasticity. At the present time, the underlying mechanism is understood to be a loss or decrease in the presynaptic Ia inhibition, which leads to an increase in exaggeration of the reflex with resultant spasticity (7). Increased muscle spasticity then leads to musculoskeletal impairment through muscle imbalance across the joints, which leads to impaired function acutely, and chronic joint contractures with skeletal deformation.

Clinical Significance
Because of the CNS injury, patients with CNS dysfunction can have significant functional impairment of their upper limbs. In the past, most evaluation and treatment of cerebral palsy centered on the lower extremities and the patient’s ability to walk. In the twenty-first century, with improved motorized wheelchair controls, public handicapped accessibility, and increased opportunities for individuals to access computers and assistive communication devices, treatment emphasis has now swung more toward maximizing functional use of the upper extremities.
History, Physical Examination, and Testing
Evaluation of children with cerebral palsy has evolved to an integration of our assessment for each aspect of manifestation of cerebral palsy: mentation, motivation, functional use patterns, sensation, static and dynamic deformities, and motor control. Evaluation of a patient’s baseline involvement in each of these areas allows formulation of realistic treatment goals and can allow measurement of treatment outcomes.
Assessment of the patient’s mentation includes evaluating the extent of musculoskeletal involvement and the degree of generalized central involvement, along with the presence or absence of seizures. Mentation can be measured by intelligence quotient testing or, for the higher functioning child, by school performance.
Motivation assessment is important for the higher functioning child, particularly with tendon transfer reconstructive surgery, which may require a highly motivated child to participate with postoperative therapy.
The child’s functional use of the hand can be quantified by using House’s classification of upper extremity functional use (Table 1). In this nine-level classification, functional use is classified into the following categories: does not use, passive assist (poor, fair, good), active assist (poor, fair, good), and spontaneous use (partial, complete). Functional use patterns can be determined through questioning the parent and child regarding use in daily activities, as well as through observation of child play (for the younger child) or the carrying out of daily activities of dressing or tying shoes (for the older child). This provides a baseline that can be used to help the physician communicate the functional goals of the surgery with the parents. The functional use can then be reassessed postoperatively by using this scale to assess for functional improvement.
Sensation can be evaluated by stereognosis, two-point discrimination, and proprioception. In our review of 40 children with spastic hemiplegia (8), we have found that stereognosis is the most sensitive discriminator of the degree of sensibility impairment, as is shown in Table 2. We found that 97% had a stereognosis impairment when using the 12 objects that are shown in Table 3. The five objects that are listed on the left column of the table discriminate gross motor function, and the seven objects that are listed on the right column of the table discriminate fine motor function. These children demonstrated recognition of 12 out of 12 objects on the unaffected side, verifying that they understood the test. Furthermore, we found that those children with severe sensibility impairment had a significant size discrepancy when the affected side was compared to the unaffected side. The shortened limb can be a useful clue to underlying sensibility deficiency, particularly in the child who is too young or too retarded to reliably perform a sensibility assessment. Children with sensibility deficiencies need to be coached to use the eyes, rather than touch, for afferent feedback. Several studies have indicated that poor sensation is not a contraindication for surgery (9). In fact, one study has reported an improvement in sensibility function after surgical intervention (10), presumably associated with increased postoperative functional use.
Level Category Description
0 Does not use Does not use
1 Poor passive assist Uses as stabilizing weight only
2 Fair passive assist Can hold object placed in hand
3 Good passive assist Can hold object and stabilize it for use by other hand
4 Poor active assist Can actively grasp object and hold it weakly
5 Fair active assist Can actively grasp object and stabilize it well
6 Good active assist Can actively grasp object and manipulate it
7 Spontaneous use, partial Can perform bimanual activities and occasionally uses the hand spontaneously
8 Spontaneous use, complete Uses hand completely, independently, without reference to the other hand
From Van Heest AE, House JH, Cariello, C. Upper extremity surgical treatment of cerebral palsy. J Hand Surg [Am] 1999;24:324, with permission.
Physical examination begins with evaluation for static deformity by examining the limb for passive range of motion of the shoulder, elbow, forearm, wrist, and hand, evaluating for joint contractures. Passive range of motion needs to be done slowly to overcome muscle tone. Contractures can exist

in the joints or the muscles, or both. Note that the finger and thumb flexor muscles are biarticular muscles, which means that they cross the wrist and finger joints. Wrist joint contractures can be checked first with the fingers flexed to assess if contractures exist in the joint itself and then with fingers extended to assess if the finger and thumb flexor muscles are contracted. If there is full passive mobility of each joint and muscle, no static deformity exists, and assessment proceeds to dynamic deformities by looking at motor function.
Stereognosis impairment No. of objects correctly identified % of children
Intact 12 3
Mild 9–11 22
Moderate 5–8 40
Severe 0–4 35
Gross Fine
Cube Safety pin
Key Paper clip
Pencil Penny
Marble Pin
Spoon Button
Rubber band
In addition to sensibility testing, the occupational therapist can be helpful in the evaluation of motor function. In the higher functioning child, the pediatric Jebsen-Taylor standardized test can be used as a baseline from which to measure the effect of subsequent treatment and also as a screen of which subtest functions have the greatest impairment (e.g., pinching small objects versus grasping large cans). Videotaping allows assessment of the spasticity that is encountered during routine activities of daily living and eliminates the stress of performance on demand in the physician’s office. Observation of the child carrying out functional tasks on videotape, as well as in the office, allows assessment of the dynamic deformities that are present. Identification of the specific spastic muscle can be determined by the joint position; for example, excessive dynamic wrist flexion with ulnar deviation identifies the flexor carpi ulnaris (FCU) as the spastic deforming force. Palpation of specific spastic muscles also localizes the source of the dynamic imbalance. Voluntary control of the muscle also needs assessment. Additionally, spasticity (increased muscle tone) versus dystonia (lack of CNS control with position of dynamic deformity, varying with time) should be noted.
Dynamic electromyography (EMG) testing is another diagnostic tool that has been used to help assess motor tone and phasic control of specific muscles (11,12). In major centers that have motion labs for gait analysis, the same technology can be used for upper extremity assessment. For example, Kozin et al. (13) have shown that biceps or brachialis spasticity, or both, can lead to elbow flexion deformity; preoperative dynamic EMG can assess spastic tone and phase control of each muscle to direct treatment to the specific offending muscles. Differences in phasic control of muscles versus continuous spastic activation can be assessed, as well as determination of central control of muscles, as children are viewed carrying out functional tasks. In our motion lab, experience shows that children need to be at least 7 years of age to appropriately cooperate with the motion lab analysis. The children have tolerated the placement of the fine needle electrodes through the use of topical anesthetic cream for 1 hour before electrode placement. The protocol that is used in our motion laboratory is demonstrated in Figure 1.
FIGURE 1. Use of the motion laboratory in assessment of upper extremity activities. The two video frames on the right show a front and side view of the child performing the pediatric Jebsen-Taylor hand test. Looking at the two angles simultaneously allows assessment of the deformity in both the sagittal and coronal planes. For example, this child can be seen to demonstrate a wrist flexion and ulnar deviation deformity. The electromyography (EMG) data on the left show 3 seconds of electrical-myographic activity of four muscles: the biceps, the pronator teres, the flexor carpi ulnaris, and the extensor carpi radialis longus/brevis. Needle electrodes are used for the pronator teres and the flexor carpi radialis. Surface electrodes are used for the biceps and the radial wrist extensor muscles. The box encompassing the central one-third of the EMG data highlights the 1 second of activity that is simultaneously shown on the video frames on the right. This motion analysis study allows for assessment of the spasticity patterns of the muscles and whether the child is able to actively control the muscles for functional use.
In summation of the evaluative process, the physician needs to integrate the results of the assessment of mentation, motivation, sensibility testing, static and dynamic deformities, and clinical and dynamic EMG motor testing to synthesize an overall treatment plan, taking into account the child’s capabilities, disabilities, and potential in the context of the child’s age. Discussion with the parents and the child is imperative in formulating the individualized treatment plan and its expected outcome.
Options and Indications
Options for treatment in cerebral palsy include occupational therapy, medications, injections, and neurosurgical procedures.
Occupational therapy includes the use of splints, stretching and strengthening programs, and active functional use activities, and many therapists may now administer electrical stimulation

programs. Two types of splints can be used: nighttime serial static splinting is used for treatment of muscle or joint contractures, and daytime splints are used for prepositioning the hand to improve active function. The indication for nighttime splinting is contracture; if no contractures of the muscles or joints exist, nighttime splinting is not necessary and is a waste of time and money for the child and family. If contractures do exist at the elbow, serial static splints can be used. If contractures exist at the wrist or fingers and thumb, a nighttime forearm-based wrist–hand orthosis may be helpful. Pronation contractures are difficult to splint and are usually treated with passive stretching. Daytime splints are usually used to pre-position the wrist in a neutral to slight “cock-up” position to help improve grasp and to preposition the thumb out of the palm to help improve pinch. If the splint is bulky or cumbersome, it interferes with rather than enhances function, thus defeating its purpose. Care should be given to ensure proper fit of the splint, so that its purpose can be achieved. Stretching and strengthening programs, along with active functional use activities, are carried out by the therapist and are taught to the parents and child as a home program. The efficacy of these treatments has been documented in limited studies (14,15 and 16), but no controlled studies have been done. Electrical stimulation is a modality that is administered by therapists as neuromuscular electrical stimulation, threshold electrical stimulation, or functional electrical stimulation. Neuromuscular electrical stimulation and functional electrical stimulation involve stimulation that produces a muscles contraction with the goal of strengthening and training a muscle. Threshold electrical stimulation involves low-intensity stimulation below the level that causes a muscle contraction and is used, while the patient sleeps, on muscles that are antagonists to the affected spastic muscles. Strengthening any muscle has been shown to improve the individual’s ability to control the muscle, but does not reduce spasticity (17). Lasting outcomes have not been reported (16).
For patients with more global tone problems, appropriate consultation or multispecialty evaluation, or both, with the rehabilitation physicians, a neurologist, or a neurosurgeon, or a combination of these, is indicated. The treatment pros and cons need to be explored for the options of tone-reducing medications (Valium, baclofen), tone-reducing injections (botulinum toxin, phenol), or tone-reducing neurosurgical interventions (baclofen pumps, selective dorsal rhizotomy), or a combination of these options. For the quadriplegic patient with overall tone control problems, medications or neurosurgical interventions should be used and stabilized before hand surgery intervention. Selective dorsal rhizotomy has been shown, in studies in the 1990s, to have an indirect tone-reducing effect on the upper extremities, in addition to its primary direct effect in the lower extremities (18,19).
For patients with more focal muscle-tone imbalance, botulinum toxin type A injections (Botox, Allergan Pharmaceuticals, Irvine, CA) have been shown to be effective in reducing spasticity in the muscles that are injected and in improving hand function (20,21,22,23 and 24). Botulinum toxin type A locally blocks the release of acetylcholine at the neuromuscular junction with a reversible action that lasts for an average of 3 to 4 months. During this period, assessment of the antagonist muscles can be made, possible surgical benefits can be assessed, antagonist muscles can be strengthened, and spastic muscles can be stretched, with the benefits lasting beyond the direct effects of the medication. For the mildly involved child, treatment with botulinum toxin type A injections may obviate the need for surgical intervention.
Author’s Preferred Treatment: Techniques, Personal Series, and Results
For patients with spastic hemiplegia that is secondary to cerebral palsy, the typical deformity includes shoulder internal rotation, elbow flexion, forearm pronation, wrist flexion with ulnar deviation, and thumb-in-palm, swan-neck, or clenched fist deformities. Optimal surgical candidates are patients who are at least 7 years of age (or mature enough to comply with postoperative regimens) with a joint-positioning deformity that interferes with active functional use of the arm. After children are diagnosed with cerebral palsy, there exists a period of, usually, 7 years before children can even be considered surgical candidates; in the meantime, children are, by definition, treated nonoperatively. Most of the nonoperative regimen revolves around parent and child education in active functional use of the affected arm to the extent that the child is able to perform.
Annual or biannual monitoring by the physician is important to assess for the development of contractures and overall developmental progress regarding age-appropriate use of the arm. For example, babies are assessed for their ability to weight bear on the arm for sitting and crawling; young children are assessed for bimanual gross motor skills, such as dressing and ball catching; school-age children are assessed for fine motor skills, such as buttoning and shoe tying; and all children are assessed for the development of contractures with growth. If failure to meet developmental milestones is encountered, a specific therapy protocol is instituted to help keep the child as close to normal developmental progress as their physical disabilities allow. Therapists are in a unique position to develop a relationship with the family and child through patient education regarding use patterns and adaptive patterns, and their services can be an integral part of the delivery of medical attention to these children.
For children with fixed deformities, nighttime static splints are prescribed, as described previously in the section Options and Indications. For children with dynamic deformities, daytime functional splints may be helpful but need to be carefully assessed as to whether they are in fact improving function rather than immobilizing and promoting patterns of the child ignoring the limb. If the daytime splints are counterproductive, they should be discontinued.
A treatment option for dynamic deformities due to spastic imbalance across a joint is botulinum toxin injection.

My series (24) of 51 children who were treated with 107 botulinum injections at an average of 10.5 years of age (with a range from 2 to 34 years of age) showed 7 of 12 quadriplegic patients exhibiting successful diminution of contractures and 32 of 39 hemiplegic and triplegic patients exhibiting improved function, as measured by range of motion, pediatric Jebsen-Taylor hand testing, and functional questionnaire evaluations. For the hemiplegic child with no fixed joint contractures, botulinum toxin injections into the FCU (for wrist flexion deformities due to a spastic FCU with active, but weak, wrist extensors) and into the pronator teres (for forearm pronation due to a spastic pronator teres with an active, but weak, supinator) are effective in improving function.
Procedures Deformity
Elbow flexion Forearm pronation Wrist flexion/ulnar deviation Finger deformity Thumb-in-palm
Soft tissue releases Biceps lengthenings (42)
Brachialis lengthenings (42)
PT releases (26)
Biceps aponeurosis releases
PQ release
FCR lengthenings (43)
Flexor pronator slides (29,30)
FCU lengthenings (43)
FDS lengthenings (43) Adductor and/or first dorsal interosseous releases (35)
First web Z-plasty
FPL lengthenings
Tendon transfers   PT rerouting (27) BR to ECRB/L (44,45)
FCR to ECRB/L (46,47)
FDS tenodesis (33)
Lateral band reroutings (31)
Spiral oblique retinacular ligament reconstruction (32)
PL to APL (34)
Bone/joint stabilization   Rotational osteotomies Wrist fusion with PRC
PRC (48)
Palmar plate capsulo-desis
PIP fusions
Distal interphalangeal joint fusions
EPL reroutings (38)
Accessory muscle of APL to EPB
MCP fusions (37)
MCP capsulodesis (36)
Interphalangeal joint fusions
APL, abductor pollicis longus; BR, brachioradialis; ECRB/L, extensor carpi radialis brevis and/or longus; ECU, extensor carpi ulnaris; EDC, extensor digitorum communis; EPB, extensor pollicis brevis; EPL, extensor pollicis longus; FCR, flexor carpi radialis; FCU, flexor carpi ulnaris; FDS, flexor digitorum superficialis; FPL, flexor pollicis longus; MCP, metacarpophalangeal joint; PIP, proximal interphalangeal joint; PL, palmaris longus; PQ, pronator quadratus; PRC, proximal row carpectomy; PT, pronator teres.
a The numbers refer to the reference for the surgical technique.
From Van Heest AE, House JH, Cariello, C. Upper extremity surgical treatment of cerebral palsy. J Hand Surg [Am] 1999;24:325, with permission.
Options and Indications
Surgical Principles
The surgical principles for treatment of spastic deformities in cerebral palsy are the following:
  • Release or lengthen the spastic or contracted muscles.
  • Augment (tendon transfers into) the weak or flaccid antagonist muscles.
  • Stabilize the joint for severe joint instability or severe fixed contractures.
The major goal in surgical reconstruction is balance across the affected joints. Surgical treatment targets joint imbalance to prevent fixed deformity and to improve functional use. Ideally, joint balance can be achieved through appropriate releases or lengthenings of the spastic muscles, with tendon transfers to augment the weak antagonist muscles, as necessary. If severe fixed deformity is already present, joint stabilization procedures may be necessary.
The surgeon is required to carefully assess the type of deformity and its treatment at each joint separately and then to synthesize them together to organize a comprehensive surgical reconstructive plan. Adequate shoulder, elbow, and forearm function are necessary for the patient to be able to appropriately position the limb in space; adequate wrist, finger, and thumb function are necessary for the patient to appropriately grasp, pinch, and release. Surgical treatment options for each of the deformities are listed in Table 4.

The most common deformity at the shoulder is internal rotation and adduction, which usually is not symptomatic, as this positions the arm effectively by the side. The most symptomatic deformity at the shoulder is external rotation and abduction. The arm interferes with normal balance or may strike objects, such as when the patient comes through a doorway. If the patient is wheelchair bound, wheelchair adaptations to restrain the arm in adduction can be tried. If conservative measures fail, a slide of the deltoid insertion for abduction deformity or an internal humeral derotational osteotomy for external rotation deformity can be performed. If the movement is primarily athetotic, surgical management is not recommended.
The most common deformity at the elbow is flexion. The primary muscles that contribute to this deformity are the biceps and the brachialis. The secondary offenders are the brachioradialis (BR) and flexor pronator wad muscles, as these muscles cross the elbow joint but are not primary elbow flexors. The primary procedure that is used to treat elbow flexion deformity is Z-lengthening of the biceps tendon with fractional lengthening of the brachialis muscle. If severe contracture exists, the BR may need to be released off its origin as well. A flexor pronator slide, as treatment for a wrist and finger flexion deformity, has a secondary effect of lessening the elbow flexion deformity.
Elbow flexion deformities of less than 45 degrees rarely functionally impair use of the limb and are usually treated nonoperatively. Elbow flexion deformities of greater than 90 degrees should be approached with caution, as the neurovascular bundle may be shortened and may limit the amount of surgical correction that is possible. Improvement of 40 degrees has been reported as an average result after biceps and brachialis lengthening (25).
The most common deformity of the forearm is pronation, which can severely limit the child’s ability to position the arm in space for grasping objects or for bringing the palms of the hands together for two-handed activities. The offending muscles are primarily the pronator teres and, secondarily, the pronator quadratus.
There are several surgical procedures to lessen forearm pronation (26,27). Two procedures that directly address the problem are pronator teres release and pronator teres rerouting. The pronator teres release works primarily through release of a deforming spastic muscle and relies on the biceps or the supinator, or both, to provide active supination after the deforming force of pronation is released. This operation is indicated if the child exhibits a severely spastic pronator teres with little control of its activity. The pronator teres is released off its insertion on the middle one-third of the radius.
The pronator rerouting procedure not only releases the pronator teres as a deforming pronation force, but also transfers the pronator back to the radius as a supinator force. This operation provides greater correction, as it releases the agonist and augments the antagonist, but care needs to be taken not to overcorrect.
Several procedures provide forearm supination as a secondary effect. The flexor pronator slide diminishes the strength of the pronator teres by releasing it off its origin. Transfer of the FCU to the extensor carpi radialis longus (ECRL) or extensor carpi radialis brevis (ECRB) provides a supination moment arm, which is greatest if the FCU is released two-thirds of the length of the forearm, as it wraps around the ulna onto the dorsum of the wrist (28). Take note that, conversely, transfer of the flexor carpi radialis (FCR) around the radius to the ECRL or brevis provides a pronation moment arm; this transfer exacerbates the forearm pronation deformity and should be avoided as a wrist tendon transfer in this population.
The most common deformity of the wrist is flexion, often with ulnar deviation, as well. This is probably the most functionally disabling deformity, as it significantly interferes with grasp and release function. Several different surgical options exist, with the choice dependent on the degree of deformity and the extent of volitional control of each muscle involved. Application of the surgical principles that were listed previously is necessary and is part of the art of designing a successful reconstructive plan: Release or lengthen the deforming spastic muscles (FCU, FCR), transfer tendons to augment the weak antagonist muscles (ECRL, ECRB), and stabilize the joint only for the severe, fixed, nonfunctioning wrist (wrist fusion).
If the wrist flexion deformity is mild, and wrist extensor control exists, weakening the wrist flexors through fractional lengthening may be sufficient. If the mild wrist flexion deformity exhibits concomitant wrist ulnar deviation, the FCU would be lengthened. If the mild wrist flexion deformity exhibits concomitant finger flexion and pronator spasticity, the entire flexor pronator mass can be lengthened by using a flexor pronator slide (29,30).
If the wrist flexor deformity is more severe, and wrist extensors are not functional, then tendon transfer surgery to augment wrist extension may be necessary. Muscles that can be transferred into wrist extensors include the BR, the extensor carpi ulnaris (ECU), or the FCU. Using the BR or the ECU as the donor tendon has the advantage of leaving both flexor tendons intact, thus avoiding overcorrection; yet, the disadvantage is not achieving balance, unless the wrist flexors are lengthened, if their spasticity is significant. Using the ECU tendon has the advantage of correction of the ulnar deviation

deformity, although this may require FCU lengthening concomitantly; yet, the disadvantage is not providing significantly more wrist extension force than is already present. Using the FCU tendon has the advantage of removing its force as a spastic wrist flexor and ulnar deviator while transferring its forces into wrist extension; yet, the disadvantage is overcorrection if the deformity is not severe or if the transfer is tensioned too tightly, particularly in the younger child.
In all cases of transfer into the wrist extensors, finger function must be assessed preoperatively with the wrist in neutral, the desired postoperative position. If the finger flexors are too tight when the wrist is brought into neutral, then a finger flexor lengthening is necessary as part of the procedure. If the patient does not have finger extensor control to allow for release of grasped objects, then a transfer into the finger extensors (extensor digitorum communis) may be indicated.
If the patient has severe wrist joint contracture that limits functional use of the hand (even as a paperweight!), consideration should be given to a proximal row carpectomy to shorten the skeletal column across the wrist joint or to a wrist fusion to hold the wrist in a fixed, more functional position. The proximal row carpectomy is used in combination with tendon transfers and releases in those cases in which passive mobility of the wrist cannot be achieved into extension. Passive mobility of the wrist is a necessary prerequisite to tendon transfer surgery, which aspires to improve active mobility, as well.
Wrist fusion predictably maintains the wrist in fixed position and is usually indicated only for improved cosmesis and use of the hand as a paperweight in the skeletally mature individual. The proximal carpal row can be removed as part of the wrist fusion to facilitate positioning of the wrist into slight extension. Wrist fusion is contraindicated in the individual who uses wrist flexion tenodesis for release function, as this function would be lost if the wrist were to be fixed in a single position.
The most common finger deformities are spastic flexion deformity and swan-neck deformity. Spastic flexion deformities are addressed in the previous section on wrist deformities, as these muscles are biarticular muscles, crossing the wrist and finger joints. Thus, they need to be lengthened in concert with the wrist flexion deformity correction, as part of the flexor pronator slide or with selective fascial lengthenings.
Swan-neck deformity of the fingers is due to dynamic imbalance of the muscles that act on the proximal interphalangeal (PIP) joint. Swan-neck finger deformity is characterized by PIP joint hyperextension with distal interphalangeal joint flexion. Swan-neck deformities can be due to a variety of causes; in cerebral palsy, swan-neck deformities are due to intrinsic muscle spasticity and often are augmented by overactivity of the extrinsic finger extensors. Many patients with cerebral palsy have better volitional control of their extrinsic finger extensors than they have of their wrist extensors. Thus, patients attempt to extend their wrists through activation of their extrinsic finger extensors. Overactivity of the extrinsic finger extensors causes metacarpophalangeal (MCP) joint hyperextension and can contribute to PIP joint hyperextension (swan-necking).
The overactivity of the extrinsic finger extensors combines with spasticity in the intrinsics to cause swan-necking. Due to cerebral palsy, the intrinsic muscles in the hand are often spastic. Spasticity of the intrinsic muscles causes overpull of the lateral bands, thus accentuating PIP joint extension. With chronic intrinsic spasticity, overpull of the lateral band causes incompetence of the transverse retinacular ligament, stretching of the PIP joint volar plate, and resultant dorsal subluxation of the lateral bands. The finger has a swan-neck deformity, as is shown in Figure 2. Excessive lengthening or surgical release of the flexor digitorum superficialis (FDS), such as is used in the superficialis-to-profundus transfer, often unmasks intrinsic spasticity that results in significant swan-neck deformities.
The indication for surgical correction of swan-neck deformities is a locking swan-neck deformity (usually greater than 40 degrees) that is not responsive to splinting and that interferes with function as part of the generalized assessment of upper limb function. For the patient with significant wrist flexion deformities and only mild swan-necking, surgical correction of wrist position alone may be adequate for treatment. For

the patient with severe swan-necking (greater than 40 degrees), rebalancing of the muscle forces at the PIP joint is necessary. Surgical options include lateral band rerouting (31), lateral band tenodesis, spiral oblique ligament reconstruction (32), intrinsic muscle slide, a resection of the ulnar nerve motor branch in Guyon’s canal (6), or superficialis tenodesis (33). The author’s preferred method is the lateral band rerouting procedure, because it requires less extensive dissection and rebalances the intrinsic and extrinsic tendons as deforming forces.
FIGURE 2. Pathophysiology of proximal interphalangeal (PIP) joint hyperextension in cerebral palsy. In cerebral palsy, the extrinsic finger extensors may be overactive in their attempt to augment wrist extension if a wrist flexion deformity exists. The intrinsic muscles are often spastic. Overactivity of the extrinsic finger extensors and spasticity of the finger intrinsics can both contribute to PIP hyperextension with resultant incompetency of the transverse retinacular ligament and volar plate. PIP hyperextension concentrates the extension forces at the PIP joint with slackening of the terminal tendon, allowing distal interphalangeal joint flexion posturing. A swan-neck deformity is characterized by PIP joint hyperflexion with distal interphalangeal joint flexion. (From Van Heest A. Lateral band re-routing in the treatment of swan-neck deformities due to cerebral palsy. Tech Hand Upper Extrem Surg 1997;1, with permission.)
The most common deformity for the thumb is in the palm. Thumb-in-palm deformity is the most complex of the deformities in the upper extremity. Treatment requires a thorough understanding of the actions of the nine muscles that act on the thumb and how these can be surgically rebalanced to provide pinch function.
Many different surgical combinations exist, with the choice dependent on the degree of deformity and the extent of volitional control of each muscle involved. Application of the surgical principles that were listed previously is necessary and part of the art of designing a successful reconstructive plan: Release or lengthen the deforming spastic muscles (flexor pollicis longus, adductor pollicis, flexor pollicis brevis), transfer tendons to augment the weak antagonist muscles (extensor pollicis brevis, abductor pollicis longus), and stabilize the joint for instability (MCP capsu-lodesis or fusion).
FIGURE 3. The type 1 thumb-in-palm deformity. Spasticity in the adductor pollicis causes significant adduction of the first metacarpal, narrowing the first web and limiting grasp. The adductor pollicis is the primary deforming force.
FIGURE 4. The type 2 thumb-in-palm

deformity. In addition to the thumb metacarpal adduction deformity from the adductor pollicis muscle spasticity, the type 2 thumb-in-palm deformity additionally has a thumb metacarpophalangeal joint flexion deformity due to flexor pollicis brevis spasticity.
Common patterns of thumb deformity have been described (34) and help determine which muscles are the deforming forces that need to be lengthened or released and which muscles are the deficient antagonists that need to be augmented.
Four types of thumb-in-palm deformity have been described. In all types, the thumb adductor is a spastic deforming force. In the type 1 thumb-in-palm deformity (Fig. 3), spasticity in the adductor pollicis causes significant adduction of the first metacarpal, narrowing the first web and limiting grasp. The adductor pollicis is the primary deforming force. In the type 2 thumb-in-palm deformity (Fig. 4), not only does adductor pollicis spasticity cause significant adduction of the first metacarpal, but also flexor pollicis brevis spasticity causes significant thumb MCP joint flexion deformity. In the type 3 thumb-in-palm deformity (Fig. 5), prolonged adductor pollicis spasticity with metacarpal adduction leads to secondary deformity, with the thumb extension and abduction through the MCP joint; this leads to secondary MCP joint instability, subluxation, and dislocation. The MCP becomes dorsally unstable with an incompetent volar plate. In the type 4 thumb-in-palm deformity (Fig. 6), in addition to the thumb metacarpal adduction deformity that is due to adductor pollicis spasticity and the MCP flexion that is due to flexor pollicis brevis spasticity, the interphalangeal joint has a flexion deformity that is due to flexor pollicis longus spasticity.
FIGURE 5. The type 3 thumb-in-palm deformity. With prolonged adductor pollicis spasticity, the thumb will extend and abduct through the metacarpophalangeal (MCP) joint (i.e., secondary MCP joint instability, subluxation, and dislocation). The MCP joint becomes dorsally unstable with an incompetent volar plate.
FIGURE 6. The type 4 thumb-in-palm deformity. In addition to the thumb metacarpal adduction deformity due to adductor pollicis spasticity and the metacarpophalangeal joint flexion due to flexor pollicis brevis spasticity, the type 4 thumb-in-palm deformity has interphalangeal joint flexion deformity due to flexor pollicis longus spasticity.
In each of these types of deformities, the offending spastic muscle needs to be released or lengthened. In the type 1 deformity, two options exist for decreasing the spastic forces of the adductor pollicis muscle, as are shown in Figures 7 and 8: the Matev (35) adductor slide or the partial adductor myotomy. In the Matev release, a palmar incision is used, and the origin of the transverse head of the adductor pollicis is elevated off the third metacarpal. The thumb is immobilized postoperatively, with the first metacarpal held in abduction, so that the adductor pollicis origin heals radially in a lengthened position. In the partial adductor myotomy, a first web Z-plasty (standard or four part) is used to increase the width of the first web skin for grasp function. Through this incision, the transverse head of the adductor pollicis is released near its insertion. The oblique head of the adductor pollicis is left intact to preserve pinch function.
In the type 2 deformity, the adductor is released in the same manner as in the type 1 deformity, and the flexor brevis is released, as well. Using the Matev incision, it is extended proximally, and the flexor pollicis brevis (FPB) origin of the thenar eminences is elevated off its origin. Using the first web incision, the FPB is released off its insertion.
In the type 3 deformity, the adductor is released in the same manner as in the type 1 deformity. Additionally, the thumb MCP joint is stabilized through a radial mid-lateral incision by using a capsulodesis technique (36) or fusion (37).
In the type 4 deformity, the adductor and FPB are released in the same manner as was described for the type 2 deformity, but the flexor pollicis longus is also lengthened in the forearm.
For the milder deformities, when antagonist control is present, a release or lengthening of the spastic muscles is indicated without additional procedures. For more severe deformities without sufficient antagonist control, tendon transfers into the abductor pollicis longus or extensor pollicis brevis are indicated. The exception is that transfers into the extensor pollicis brevis are not indicated for the type 3 thumb-in-palm deformity, unless the joint is fused, as a strong MCP extensor exacerbates the MCP extension instability.
Tendon transfers to augment extension and abduction of the thumb include transfers into the abductor pollicis longus and extensor pollicis brevis or rerouting of the extensor pollicis longus (EPL) into the first dorsal compartment. Donor tendons for transfer include the FCR (if the FCU is not transferred as part of the wrist correction), the

BR, and the palmaris longus. If satisfactory tendon donors are not available, tenodeses can be carried out in the lower-functioning hand. Rerouting of the EPL from the third dorsal compartment, where it acts as a secondary thumb adductor, into the first dorsal compartment, converts it into a thumb extensor abductor (38).
FIGURE 7. Type 1 thumb-in-palm deformity Matev release (35). Through a palmar incision, the origin of the transverse head of the adductor pollicis is elevated off the third metacarpal. The thumb is immobilized postoperatively with the first metacarpal held in abduction so that the adductor pollicis origin heals radially in a lengthened position.
FIGURE 8. Thumb-in-palm partial adductor myotomy. Through a first web Z-plasty incision, used to increase the width of the first web skin for grasp function, the transverse head of the adductor pollicis is released near its insertion. The oblique head of the adductor pollicis is left to preserve pinch function.
Results and Outcome, Review of the Literature, and Factors That Affect Outcome
A review of the literature documents overall functional improvement of children with spastic hemiplegia after operative treatment (9,39,40). Nylanders (39) documented that functional improvement was achieved by 6 months postsurgery and was maintained at a 4.5-year follow-up in 24 children. Eliassion (40) reported on 32 children who were treated with tendon transfers and muscle releases and showed functional improvement in all children regardless of the degree of preoperative impairment. He noted that the extent of improvement was based on preoperative functional level. I reported on 134 patients who were treated with soft tissue release, tendon transfer, and joint stabilization procedures (9). Most commonly, all deformities were corrected during one operation, with an average of four procedures performed per surgery. Surgical results showed an average improvement of 2.6 level according to House’s classification of upper extremity functional use (Table 1). This functional improvement was not affected by level of mentation, two-point discrimination, stereognosis function, or type of cerebral palsy. (Note that only three athetoid patients were treated surgically over 25 years.) Patients with poor motor control did have less improvement in functional use. Most patients who were treated surgically were highly motivated.
Author’s Preferred Treatment: Techniques, Personal Series, and Results
The challenge of upper extremity reconstructive surgery for spastic conditions is that there is no single surgical solution that is applicable to all patients. Each patient has varying degrees of fixed contractures and varying degrees of spasticity that causes dynamic deformity. Spasticity causing muscle imbalance across joints limits function to the greatest extent in the wrist and in the thumb. Thus, surgical planning usually targets the wrist and thumb, as surgical reconstruction of these joints provides the greatest functional improvement.

For the child younger than 7 years old or uncooperative for tendon transfer surgery, I prefer nonoperative management as discussed above. For the child older than 7 years old and cooperative, this is an optimum time for surgical intervention, before the development of fixed contractures.
In my practice, I would recommend a motion laboratory analysis before surgical intervention for tendon transfer surgery. At the present time, I use fine needle electrodes in the pronator teres and FCU muscles (Fig. 1). If either muscle shows significant spasticity with continuous electrical activity and no evidence of selective control, then I would recommend a release or lengthening of the muscle. If either muscle shows appropriate phasic control of its electrical activity, then I would recommend tendon transfer using that muscle. For example, if the pronator teres is continuously spastic, then I would perform a pronator teres release; if the pronator teres is under phasic control, then I would perform a pronator teres rerouting. Another example, if the FCU is continuously spastic, then I would perform an FCU lengthening and transfer the BR or ECU for wrist extension; if the FCU is under phasic control, then I would perform an FCU tendon transfer for wrist extension.
Surgical Technique
For tendon transfer surgery, including the elbow, the patient is positioned supine with a sterile tourniquet to allow for tourniquet removal before tensioning of the biceps lengthening. For all other forearm, wrist, and hand surgeries, a standard tourniquet is used.
For biceps and brachialis lengthening, an anterior “lazy S” incision is used across the antecubital fossa with the proximal limb extending medially and the distal limb extending laterally toward the biceps insertion. After identification and protection of the lateral antebrachial nerve, the biceps tendon is isolated, and the lacertus fibrosis is released. The biceps tendon is lengthened using a step-cut Z-lengthening technique. The brachialis is lengthened using a series of transverse incisions through the fascial layers. The biceps is repaired in a lengthened position at the conclusion of surgery when the sterile tourniquet is deflated.
For the pronator teres release or rerouting, a longitudinal radial incision is used along the middle one-third of the radius. Dissection is carried down onto the radial insertion of the pronator teres, with careful protection of the radial artery and superficial radial nerve. For the pronator teres release, the tendon is released along its radial insertion and its fascial attachments to the extent that retraction of the muscle can be verified. For pronator teres rerouting, the muscle is freed off its volar insertion with a long periosteal extension and tagged with 0-0 nonabsorbable suture. A window is then made in the interosseous membrane of sufficient size and position to pass the pronator teres. The pronator teres is passed through the window, dorsally around the radius, back onto its insertion, and, if tendon length allows, back onto the volar aspect of the radius to provide maximum supination (41). The tendon is sewn firmly onto periosteum or attached with bone anchors or bone drill holes. The arm is immobilized postoperatively for both procedures in maximum supination.
For the wrist tendon transfers, the FCU is transferred to the ECRB (Fig. 9A) through a long ulnar-sided incision. Dissection is carried down along the FCU insertion on the pisiform with careful identification and protection of the ulnar nerve and artery. The FCU is transected at its pisiform insertion and tagged with a 2-0 nonabsorbable grasping suture. The muscle is mobilized off its ulnar origin and fascial investments over the distal two-thirds of the ulna to maximize its excursion (Fig. 9B). The most distal ulnar motor nerve branch to the FCU can be sacrificed if necessary to facilitate excursion. A dorsal incision is then made over the ECRB just proximal to the extensor retinaculum. A fascial window is cut over the ECRB just distal to the thumb outcropper muscles. A subcutaneous tunnel is then made from the proximal end of the ulnar incision to the dorsal radial incision, and the FCU tendon is passed around the ulnar border of the forearm (Fig. 9C). Three Pulvertaft weaves of the FCU end-to-side into the ECRB tendon are then made and sewn in place with tensioning so that at rest, in a gravity-dependent position, the wrist sits at neutral (Fig. 9D). Passive elongation of the muscle should allow for wrist flexion, and active firing of the muscle should allow for wrist extension. After the wounds are closed, the wrist is immobilized in slight wrist extension.
For the fingers and thumb, there is no “cookbook” formula that can be used for the spastic hand. Use of the surgical principles outlined above is imperative with customization to the reconstruction plan and the degree of spasticity and/or deformity. In the fingers, the superficialis to profundus transfer for finger flexion deformity is used for the nonfunctional hand to help with hygiene problems. A volar forearm incision is used with identification of the FDS superficially, protection of the median nerve in the intermediate layer, and identification of the flexor digitorum profundus (FDP) in the deep layer. A 2-0 nonabsorbable suture is used to sew the four FDS tendons together in their resting cascade, and the tendons are then transected at their distal-most end, just proximal to the carpal tunnel, with the fingers in full flexion to deliver the most amount of tendon into the wound. A 2-0 nonabsorbable suture is used to sew the FDP tendons together in their resting cascade, and the tendons are then transected at the proximal end, just at the musculotendinous junction. The distal end of the FDS tendons are then woven into the proximal ends of the FDP tendons using 2-3 Pulvertaft weaves on each of the four fingers’ tendons. Care is taken to maintain the normal cascade and tension so that the hand at rest lays flat.
For the thumb, I prefer to lengthen the adductor pollicis muscle through a first web Z-plasty incision (Fig. 8), using


a four-part Z-plasty for the more severely contracted skin. A complete release of the fascia and a partial release of the transverse head allow first metacarpal abduction without precluding pinch. If the metacarpal joint is dorsally unstable, I prefer an MCP capsulodesis. A radial midlateral incision is made with dissection carried directly to an MCP capsulotomy. The volar plate insertion is usually attenuated. Both the metacarpal neck and the sesamoids are abraded. Either a bone suture anchor or a periosteal stitch is used to bring the sesamoids into direct contact with the metacarpal neck. The MCP joint should sit in 20 to 30 degrees of flexion, and should come to neutral with passive stretch. The joint is then pinned in 20 to 30 degrees of flexion. If the extensors or abductors are weak, augmentation is performed through tendon transfer. If the EPL is well controlled, a transfer of the EPL from the third compartment into the first compartment with shortening of the tendon, using the dorsal wrist incision from concomitant wrist tendon transfer, is my preferred technique.
FIGURE 9. The flexor carpi ulnaris (FCU) to extensor carpi radialis brevis (ECRB) tendon transfer. A: For the severe wrist flexion deformity, the FCU can be released as a deforming force and transferred to the ECRB to augment weak or absent wrist extension. B: Through a long ulnar-sided incision, the FCU is harvested from its pisiform insertion and tagged with a 2-0 nonabsorbable grasping suture. The muscle is mobilized off its origin and fascial investments over the distal two-thirds of the ulna to maximize its excursion. C: The FCU tendon is passed through a subcutaneous tunnel from the ulnar incision into a dorsal incision over the ECRB, just proximal to the extensor retinaculum. Three Pulvertaft weaves are sewn in place with 2-0 nonabsorbable sutures. D: The transfer is tensioned so that at rest, in a gravity-dependent position, the wrist sits at neutral. Passive elongation of the muscle should allow for wrist flexion, and active firing of the muscle should allow for wrist extension. ECRL, extensor carpi radialis longus.
I advocate multiple simultaneous upper extremity procedures for multilevel correction of the spastic limb in cerebral palsy. Postoperatively, a well-padded cast that allows for swelling but maintains intraoperative positioning is applied. The cast is removed 4 weeks later, and the patient is fitted with a custom Orthoplast splint. The splint is worn full-time for 4 weeks, with the patient removing it for active range of motion, light at-table activities, and hygiene. After 4 weeks of full-time wear, if the patient is maintaining the position of joint correction and learning active range of motion, then the patient is progressed to part-time use protection during sleep and activities (e.g., school and sports) and is started on strengthening exercises with gentle passive range of motion if needed for mobility.
After this 3-month postoperative program, a long-term upper extremity functional use protocol, encouraging long-term bimanual skills, is recommended.
Algorithm for Evaluation and Management (Based on the Author’s Preferred Treatment)
Based on the author’s preferred treatment, the algorithm outlined in Figure 10 has been developed.
FIGURE 10. Algorithm of author’s preferred treatment. APL, abductor pollicis longus; BR, brachioradialis; C.P., cerebral palsy; ECRB, extensor carpi radialis brevis; ECU, extensor carpi ulnaris; EPB, extensor pollicis brevis; EPL, extensor pollicis longus; FCR, flexor carpi radialis; FCU, flexor carpi ulnaris; FDP, flexor digitorum profundus; FDS, flexor digitorum superficialis; FPB, flexor pollicis brevis; FPL, flexor pollicis longus; MCP, meta-carpophalangeal; O.T., occupational therapy; P., pronator; PL, palmaris longus.

All surgical procedures carry risk, which must be weighed against the potential benefits that most commonly are achieved. Multiple simultaneous upper extremity procedures can be safely performed with meticulous attention to pre-, intra-, and postoperative details.
Preoperatively, patients must be screened for anesthetic complications, including a bleeding screen for patients on long-term Depakote antiseizure medications; screening for bladder and lung infections, particularly for patients with poor urinary or pulmonary control; and assessment of nutritional status (height and weight percentiles for age).
Intraoperative attention to wound care is imperative to avoid wound healing problems, particularly in Z-plasty approaches. Large wounds should be treated with a postoperative drain to prevent hematoma formation. Nerve and artery injury due to overzealous correction of joint position is to be avoided.
Postoperatively, the splint or cast should be adequate to allow for postoperative swelling and should be split if excessive swelling is encountered. Many children with spasticity do not have a normal preoperative sensory or motor examination and may not have normal mentation, so normal parameters cannot be used to monitor for compartment syndrome. Premature removal of the cast or splint, as well as overzealous patient activities, can lead to tendon rupture or attenuation. Excessive immobilization can lead to excessive adhesion formation, diminishing the eventual functional use.
Long-term problems most commonly center around loss of the balance achieved at the time of the surgery. Many children have tendon transfers as young as 7 years old and, with continued skeletal growth, may have recurrent deformity. Overcorrection can also occur with the “opposite” deformity occurring. Additionally, further “fine tuning” surgery may be necessary to address complications that develop after correction of the original deformity. For example, after corrective lengthening of the finger and wrist extensors, intrinsic spasticity in the hand may be unmasked. This may be assessed through an ulnar nerve motor block in Guyon’s canal; if successful, an ulnar motor neurectomy can be performed. Recurrent over- or undercorrection can be treated with stretching, splinting, and nonoperative measures as outlined in Nonoperative Treatment.
Principles that help prevent these complications include
  • Do not overcorrect deformity, particularly in the younger child.
  • Leave options to reverse the surgical correction if necessary.
  • Keep functional grasp and release as your highest priority in your surgical planning.
  • Avoid wrist arthrodesis as this precludes the tenodesis effect of the wrist for finger use.
Surgical expectations should be realistically discussed with your patients and their parents to avoid unrealistic expectations. In general, using the functional use scale in Table 1, an average of two and one-half functional levels of improvement can be achieved with surgical intervention. Most commonly, this would mean that a child who presents with a severely flexed wrist with a thumb-in-palm deformity that limits use of the limb to a good passive assist could be improved to a fair to good active assist with better wrist and thumb positioning.
As surgeons, we must remember that we are treating only the secondary peripheral manifestations of a primary CNS dysfunction that persists.
1. Minear WL. A classification of cerebral palsy. Pediatrics 1956;18:841.
2. Dunin-Wasowicz D, Rowecka-Trzebicka K, Milewska-Bobula B, et al. Risk factors for cerebral palsy in very low birth weight infants in the 1980s and 1990s. J Child Neurol 2000;15:417–420.
3. Fuchs Z, Blumstein T, Novikov I. Morbidity, comorbidity, and their association with disability among community-dwelling oldest-old in Israel. J Gerontol A Biol Sci Med Sci 1998;53:447–455.
4. Lackland DT, Bachman DL, Carten TD. The geographic variation in stroke incidence in two areas of the southeastern stroke belt: the Anderson and Pee Dee Stroke Study. Stroke 1998;29:2061–2068.
5. Sacco RL, Boden Albala B, Gan R. Stroke incidence among white, black and Hispanic residents of an urban community: the Northern Manhattan Stroke Study. Am J Epidemiol 1998;147:259–268.
6. Trumble TE, Van Heest A. Stroke, traumatic brain injury, and cerebral palsy. In: Trumble TE, ed. Principles of hand surgery and therapy. Philadelphia: WB Saunders, 2000:361–375.
7. Katz R. Presynaptic inhibition in humans: a comparison between normal and spastic patients. J Physiol 1999;93:379–385.
8. Van Heest AE, House J, Putnam M. Sensibility deficiencies in the hands of children with spastic hemiplegia. J Hand Surg 1993;18:278–281.
9. Van Heest AE, House JH, Cariello C. Upper extremity surgical treatment of cerebral palsy. J Hand Surg 1999;24:323–330.
10. Dahlin LB, Komoto-Tufvesson Y, Salgeback S. Surgery of the spastic hand in cerebral palsy. Improvement in stereognosis and hand function after surgery. J Hand Surg 1998;23:334–339.
11. Hoffer MM. The use of the pathokinesiology laboratory to select muscles for tendon transfers in the cerebral palsy hand. Clin Orthop 1993;288:135–138.
12. Hoffer MM, Perry J, Melkonian G. Postoperative electromyographic function of tendon transfers in patients with cerebral palsy. Dev Med Child Neurol 1990;32:789–791.

13. Kozin SH, Keenan MH. Using dynamic electromyography to guide surgical treatment of the spastic upper extremity in the brain-injured patient. Clin Orthop 1993;288:109–117.
14. Nogen AG. Medical treatment for spasticity in children with cerebral palsy. Child Brain 1976;2:304–308.
15. Hines AE, Crago PE, Villian C. Functional electrical stimulation for reduction of spasticity in the hemiplegic hand. Biomed Sci Instrum 1993;29:259–266.
16. Carmick J. Clinical use of neuromuscular electrical stimulation for children with cerebral palsy. Part II: upper extremity. Phys Ther 1993;73:514–527.
17. Steinbok P, Reiner A, Kestle JR. Therapeutic electrical stimulation following selective posterior rhizotomy in children with spastic diplegic cerebral palsy: a randomized clinical trial. Dev Med Child Neurol 1997;39:515–520.
18. Beck AJ, Gaskill SJ, Marlin AE. Improvement in upper extremity function and trunk control after selective posterior rhizotomy. Am J Occup Ther 1993;47:704–707.
19. Loewen P, Steinbok P, Holsti L, et al. Upper extremity performance and self-care skill changes in children with spastic cerebral palsy following selective posterior rhizotomy. Pediatr Neurosurg 1998;29:191–198.
20. Fehlings D, Rang M, Glazier J, et al. An evaluation of botulinum-A toxin injections to improve upper extremity function in children with hemiplegic cerebral palsy. J Pediatr 2000;137:300-303,331–337.
21. Autti-Ramo I, Larsen A, Peltonen J, et al. Botulinum toxin injection as an adjunct when planning hand surgery in children with spastic hemiplegia. Neuropediatrics 2000;31:4–8.
22. Wall SA, Chait LA, Temlett JA, et al. Botulinum A chemodenervation: a new modality in cerebral palsied hands. Br J Plast Surg 1993;46:703–706.
23. Van Heest AE. Applications of botulinum toxin in orthopaedics and upper extremity surgery. Tech Hand Upper Extrem Surg 1997;1:27–34.
24. Van Heest AE. A prospective evaluation of treatment of the upper extremity in spastic hemiplegia using botulinum toxin. In: Ogino T, ed. Congenital differences of the upper limb. Kyoto: Yamagata University School of Medicine; 2000:307–320.
25. Mital MA, Sakellarides HT. Surgery of the upper extremity in the retarded individual with spastic cerebral palsy. Orthop Clin North Am 1981;12:127–141.
26. Strecker WB, Emanuel JP, Dailey L, et al. Comparison of pronator tenotomy and pronator rerouting in children with spastic cerebral palsy. J Hand Surg 1988;13:540–543.
27. Sakellarides HT, Mital MA, Lenzi WD. Treatment of pronation contractures of the forearm in cerebral palsy by changing the insertion of the pronator radii teres. J Bone Joint Surg 1981;63:645–652.
28. Van Heest AE, Murthy NS, Sathy MR, et al. The supination effect of tendon transfer of the flexor carpi ulnaris to the extensor carpi radialis brevis or longus: a cadaveric study. J Hand Surg 1999;24:1091–1096.
29. Inglis AE, Cooper W. Release of the flexor-pronator origin for flexion deformities of the hand and wrist in spastic paralysis. J Bone Joint Surg 1966;48:847–857.
30. White WF. Flexor muscle slide in the spastic hand: the Max Page operation. J Bone Joint Surg 1972;54:453–459.
31. Tonkin MA, Hughes J, Smith KL. Lateral band translocation for swan-neck deformity. J Hand Surg 1992;17:260–267.
32. Littler JW. The finger extensor mechanism. Surg Clin North Am 1967;47:415–432.
33. Swanson AB. Surgery of the hand in cerebral palsy and the swan neck deformity. J Bone Joint Surg 1960;42:951–964.
34. House J, Gwathmey F, Fidler M. A dynamic approach to the thumb-in-palm deformity in cerebral palsy. J Bone Joint Surg 1981;63:216–225.
35. Matev I. Surgery of the spastic thumb-in-palm deformity. J Hand Surg 1991;16:346–348.
36. Filler BC, Stark HH, Boyes JH. Capsulodesis of the meta-carpophalangeal joint of the thumb in children with cerebral palsy. J Bone Joint Surg 1976;58:667–670.
37. Goldner JL, Koman LA, Gelberman R, et al. Arthrodesis of the metacarpophalangeal joint of the thumb in children and adults: adjunctive treatment of thumb-in-palm deformity in cerebral palsy. Clin Orthop 1990;253:75–89.
38. Manske PR. Redirection of extensor pollicis longus in the treatment of spastic thumb-in-palm deformity. J Hand Surg 1985;10:553–560.
39. Nylanders G, Carlstrom C, Adolfsson L. 4.5 year follow-up after surgical correction of upper extremity deformities in spastic cerebral palsy. J Hand Surg 1999;24:719–723.
40. Eliassion AC, Ekholm C, Carlstedt T. Hand function in children with cerebral palsy after upper-limb tendon transfer and muscle release. Dev Med Child Neurol 1998;41:284–285.
41. Van Heest AE, Sathy M, Schutte L. Cadaveric modeling of the pronator teres rerouting tendon transfer. J Hand Surg 1999;24:614–618.
42. Mital MA. Lengthening of the elbow flexors in cerebral palsy. J Bone Joint Surg 1979;61:515–522.
43. Zancolli EA. Structural and dynamic bases of hand surgery, 2nd ed. Philadelphia: Lippincott Williams & Wilkins, 1968.
44. House JH, Gwathmey FW. Flexor carpi ulnaris and the brachioradialis as a wrist extension transfer in cerebral palsy. Minn Med 1978;61:481–484.
45. McCue FC, Honner R, Chapman WC. Transfer of the brachioradialis for hands deformed by cerebral palsy. J Bone Joint Surg 1970;52:1171–1180.
46. Green WT. Tendon transplantation of the flexor carpi ulnaris for pronation-flexion deformity of the wrist. Surg Gynecol Obstet 1942;75:337–342.
47. Green WT, Banks HH. Flexor carpi ulnaris transplant and its use in cerebral palsy. J Bone Joint Surg 1962;44:1343–1352.
48. Omer GE, Capen DA. Proximal row carpectomy with muscle transfers for spastic paralysis. J Hand Surg 1976;1:197–204.