13.5 Upper limb management in cerebral palsy

The upper extremity is involved in all varieties of cerebral palsy. These limbs display the ‘positive’ features of the upper motor neuron syndrome—spasticity, hyper-reflexia, clonus, and co-contraction—but they also demonstrate the negative features including weakness, sensory deficit, poor selective, motor control (see Chapter 13.3, Table 13.3.2), as well as dystonia, which further influence the fine motor skills. Upper limbs perform a multitude of tasks and must work together, although they have differing functions depending on dominance. Whereas children need both lower limbs to walk, many upper-limb tasks can be performed single-handedly, and neglect or ‘learned non-use’ of the other hand can be a problem. Management of the upper limb in cerebral palsy involves detailed assessment by a multidisciplinary team (see Chapter 13.3, Table 13.3.4) and careful planning.

Cerebral palsy is classified by the predominant type of movement disorder: spasticity is present in 85% of those affected and is the most amenable to treatment. It is also classified by topographical involvement. Those with total body involvement have the greatest problems with motor control, spasticity, sensory deficit, and contractures leading to poor function. Goals of treatment are usually to improve hygiene and ease of dressing for these patients. Spastic diplegic patients have meagre involvement of the upper extremity and so rarely require intensive upper limb therapy or surgery. Hemiplegic patients can be helped most in terms of function and cosmesis.

A careful and repeated evaluation should be performed evaluating the factors listed in Box 13.5.1.

Sensory status is a predictor of the degree of spontaneous use of the limb.

Light touch, pain, and temperature sensations are present in cerebral palsy, but whether they are ‘normal’ is difficult to determine. They are ‘normal’ for that patient who will report them as such. Proprioception is measured objectively and tends to be more affected distally. Stereognosis is the child’s ability to recognize objects by feel, and its reduction correlates with a decrease in limb size. Graphesthesia is the recognition of pictures or letters drawn on the palm. Two-point discrimination is considered satisfactory if less than 5–10mm.

Resting posture gives an idea of the amount of spasticity present. It tends to cause shoulder adduction and internal rotation, elbow flexion, forearm pronation, and wrist flexion. Fingers are flexed or show swan-neck deformity. The thumb may be adducted, or adducted and flexed, the so called ‘thumb-in-palm’ deformity. Stretching of the affected muscles meets with resistance which is velocity dependent and correlates with the degree of spasticity present. This can be assessed further with the Modified Ashworth and the Modified Tardieu Scales.

Spasticity often predominates in the shoulder adductors, elbow flexors, pronators, wrist, and finger flexors, and these muscles often overpower their antagonist muscles. The wrist and finger extensors, supinators, and abductor pollicis longus can be weak with poor voluntary control. Assessment of their power may only be possible after spasticity in other muscle groups has been reduced (by motor blocks or botulinum toxin A).

Spasticity, fibrous contractures within spastic muscles, and joint contractures can be difficult to distinguish. It is usually possible to overcome spasticity, unless severe. Fibrous contractures cannot be overcome by passive movement, but can sometimes be distinguished from joint contractures by altering joint positions when testing muscles that cross more than one joint. For example, if finger flexors are tight, passive finger extension may only be possible with wrist flexion. Joint contractures such as elbow and wrist flexion and a pronation contracture due to shortening of the intraosseous membrane may develop in adolescence or adulthood. Occasionally the radial head may dislocate.

A number of objective, reproducible, validated tests, such as the Melbourne Unilateral Upper Limb Assessment, the Quality of Upper Extremity Skills Test (QUEST), the Assisting Hand Assessment (AHA), and the Shriners Hospital Upper Extremity Evaluation (SHUEE) can help assess and score the child’s functional ability. Functional tests should be recorded on video and form a useful outcome measure after surgery. Bi-manual tests which measure what the child actually does rather than what he or she is capable of (the performance gap) will give a better measure of actual function. The chosen test needs to be responsive to change in order to be a good measure of outcome. In addition, the Manual Ability Classification System (MACS) reports the child’s ability to handle objects in daily life. It does not assess the hands independently of each other. It gives a score of I–V and correlates well with Gross Motor Function Classification System (GMFCS) level ratings. The House score assesses the affected limb (Table 13.5.1).

Wrist flexion and thumb deformities have been shown to be major contributors to functional problems and in fact the House score was originally used to assess function before and after thumb surgery (Figure 13.5.1).

Three-dimensional motion analysis for the upper limb is in its infancy and not yet standardized between different centres. Electromyography may help determine which muscles are used in grasp and which in release, but is not widely used. Radiographs are rarely necessary.

The treatment goals must be realistic and clearly defined by patient, carers, therapists, and surgeon together. In the severely affected quadriplegic patient, the goals may be to improve hygiene and ease of dressing. In many hemiplegic patients it is possible to improve function. Older children may simply seek an improvement in appearance but often do not admit to this unless directly questioned. Treatment must be individualized.

A number of treatments are available. Splinting to serially stretch contractures or to position the limb to its biomechanical advantage; stretching spastic muscles or strengthening weak ones; neurodevelopmental therapy to encourage functional movement patterns and inhibit primitive posturing; conductive education which practises motor skills for functional use and constraint induced movement therapy to encourage use and awareness of the more affected limb by restricting the more able hand, in much the same way as eye patches are used over a good eye to help train the ‘lazy’ eye.

Botulinum toxin blocks the presynaptic release of acetyl choline at the neuromuscular junction. The dose is calculated according to the patient’s bodyweight, and each target muscle should be located with a nerve stimulator or ultrasound probe to ensure accurate drug placement. Splinting and stretching as well as strengthening programmes for weak antagonists should be implemented following injections.

Many muscles of the upper limb cross more than one joint and therefore the limb must be considered as a whole and not as individual joints in isolation (Table 13.5.2). For example, a flexor–pronator slide will release the tight pronator, finger and wrist flexors, but will also slightly improve elbow extension. A flexor carpi ulnaris to extensor carpi radialis brevis transfer will augment wrist extension, but will also significantly increase active supination. Equally, a wrist should not be arthrodesed in a neutral position without addressing tight finger flexors or the patient will no longer be able to release their grasp. It is also important to remember that the primary issue is with brain function and not to see the problem as a purely biomechanical one. A neglected limb is unlikely to gain function despite surgery. Generally, the surgeon has the following options for treatment:

Motor nerves may be divided, partially resected at the neuromuscular junction, or injected with phenol.

If possible, all the applicable procedures should be performed in one sitting (multilevel surgery for the upper limb).

Postoperatively the arm is immobilized in a cast, the position dependent on which procedures have been carried out. This is maintained for 6 weeks and then replaced with a thermoplastic splint to be worn at night time for 3 months or more to prevent recurrence of contractures. Movement is commenced on the non-immobilized joints immediately postoperatively, and therapy is intensified once the cast comes off. Occupational and physiotherapy are essential adjuncts to medical and surgical treatment.

A systematic review of the literature on BTA identified 12 studies of high methodological quality. Six of ten that measured spasticity showed improvement; three of ten showed increased range of movement and six of ten showed improvement in function. The authors concluded, however, that due to differences in treatment goals, invalid assessment instruments and insufficient statistical power there was insufficient evidence to state that injections were beneficial.

Although there are no prospective randomized controlled trials assessing the efficacy of surgical treatment for the upper limb in cerebral palsy, from the large retrospective studies, the literature would seem to support surgery for improvement in function, cosmesis, and hygiene. A review of the literature from 1966–2006 concluded that surgery improved the position of the hand and there are indications that it might improve hand function.

Prospective studies using validated outcome measures and long-term follow-up of treated patients, as well as the development of valid three-dimensional upper-limb modelling, will enhance our understanding of the effect of interventions on the upper limb in cerebral palsy.

Management of the upper limb in cerebral palsy requires assessment by a multidisciplinary team with an understanding of neuromuscular disability and of anatomy. Realistic goals need to be agreed with the patient and family, and validated outcome measures need to be documented. This is an exciting field for clinicians and researchers, with ample scope for future development and research.