Near target muscle
Pure motor fibers
As large a number of motor axons as possible
Synergistic to target muscle
Tension free transfer through full range of motion
Divide “donor distal” and “recipient proximal” to allow tension free nerve approximation
If redundancy exists, trim recipient nerve to minimize distance from regenerating donor axons to target muscle motor endplates
Appropriate therapy for motor re-education
Nerve injuries in the upper extremity frequently cause permanent disability with devastating consequences on daily living and the ability to work. Direct primary repair of a nerve transaction is advocated if possible and appropriate for the mechanism of injury. Large nerve gaps, proximal injuries with crush, transaction-, and avulsion-type injuries prohibit direct repair and require secondary nerve reconstruction after acute management and stabilization. Until the last century, the results of nerve repair in the extremity and especially of the brachial plexus were viewed with pessimism. In the latter half of the 20th century, advances in peripheral nerve surgery, including improvements in nerve repair and grafting techniques, and improved knowledge of internal topography, injury pattern, and regenerative ability of the peripheral nerve have contributed to better outcomes.
The level, mechanism, and severity of the injury determine the course of management and the timing of surgical procedures. Factors such as muscle atrophy after denervation and degeneration of the neuromuscular junction limit the opportunity for nerve reconstruction and adversely influence the outcome. If nerve fibers do not successfully reinnervate the target muscle within approximately 1 to 1.5 years of injury, the motor end plates disappear, the muscle architecture is destroyed, and muscle fibers are eventually replaced by fat with time. Therefore, “time is muscle,” and with complex injury patterns it is imperative to closely follow the examination and results of diagnostic tests to make a clear diagnosis in a timely fashion such that reinnervation can occur before it is too late.
Advances in nerve repair and in the understanding of the internal topography of the nerve have contributed to the development of nerve transfers. Nerve-to-nerve transfers offer a superior alternative for functional restoration in isolated or multiple-nerve injuries when early reinnervation of the target end-organ is necessary, such as in proximal injuries or in delayed treatment. Expendable sensory or motor axons close to the end-organ allow for earlier regeneration and preclude the need for nerve grafts. Nerve transfers have traditionally been advocated for otherwise unsalvageable root avulsion injuries of the brachial plexus in which reconstruction with nerve grafts is suboptimal or not possible. Nerve graft reconstruction of such proximal injuries requires regeneration over long distances before reaching target end-organs, frequently resulting in suboptimal functional recovery. The patient population with brachial plexus injury provided the stimulus for the application of novel reconstructive options that would offer faster reinnervation of muscles to minimize atrophy and degeneration of the motor end plate. Consequently, nerve transfers became a routine part of the early surgical management of upper brachial plexus injuries with very good outcomes. The experience with tendon transfers and an understanding of the principles of motor reeducation set the stage for the use of nerve transfers as an alternative to muscle or tendon transfers. The advantages typically include shorter operative time, shorter duration of morbidity and recovery, and faster target muscle reinnervation with less atrophy and motor end-plate fibrosis.
This writer has progressively applied this technique to more distal and isolated nerve injuries in the forearm and hand with encouraging results. As our understanding of the innervation patterns of muscles and the internal topography of peripheral nerves increases, our application of nerve transfers has become more widespread in the management of a range of peripheral nerve injuries, including those with more distal and isolated nerve trauma and those for which tendon transfers are conventionally recommended. My practice has evolved toward a broader application of distal nerve transfers to minimize operative time, morbidity, and recovery and to more quickly reinnervate target organs to optimize the functional outcome.
Principles of Nerve Transfer
Nerve transfers involve repair of the proximal aspect of a functioning motor nerve to the distal aspect of a nonfunctioning motor nerve of the target muscle. While many of the principles for nerve transfers are common to tendon transfers, there are unique considerations as well. Redundant and expendable nerve fascicles or branches of the donor nerve are utilized to minimize morbidity and the possibility of downgrading function. As with tendon transfers, a donor nerve that supplies muscles synergistic to the target muscle is preferred to facilitate postoperative therapy and motor reeducation. While a nerve supplying a nonsynergistic, or even antagonistic, muscle group may be used, the rehabilitation in such cases is more difficult and less intuitive, and functional recovery may be less optimal. Biomechanical considerations of muscle type and amplitude, vector, and tension that affect strength of a transferred muscle, do not apply, since muscles reinnervated by nerve transfers maintain their anatomic location and attachments. Other advantages of a nerve transfer include (1) the capacity to restore sensibility in addition to motor function, (2) the possibility of restoring multiple muscle groups with a single nerve transfer, and (3) the avoidance of dissection and scarring of the muscle bed that may limit excursion and strength. If known, the donor nerve should have a similar number of nerve fibers as the recipient nerve. Intraplexal transfers may also be more effective than those that use extraplexal nerve donors, generally because of proximity to the target muscles. The most critical factor for the success of a nerve transfer is the quality of the donor motor nerve in terms of axon count.
As with any peripheral nerve surgery, nerve transfer for restoration of motor function demands a timely approach. If nerve fibers do not reach the motor end plate within, ideally, 1 year of injury, the muscle will not work. Thus, an additional advantage of nerve transfer surgeries is that the transfer does not need to be performed in the zone of injury but should be done close to the recipient nerve motor end plate. This in essence shifts a proximal-level nerve injury to a distal-level injury and avoids dissection in the original area of injury, which may have significant scarring—thus facilitating operative dissection. Sensory reinnervation, on the other hand, can be performed at any time after injury. At the end of this section, specific operations and options for delayed restoration of motor function, with free-functioning muscle transfers, will be discussed.
Nerve transfers may also be used as an alternative to nerve grafting or primary repair in certain cases because of the ability to convert a proximal high-level nerve injury to a low-level nerve injury. For example, a high radial nerve injury may be treated by transfer of redundant median nerve branches [to the flexor digitorum superficialis (FDS) and palmaris longus (PL)] to distal radial nerve branches to reinnervate the extensor carpi radialis brevis (ECRB) and the muscles innervated by the posterior interosseus nerve. The transfer is performed at the proximal forearm, allowing more timely restoration of wrist and finger extension. Proximal ulnar nerve injuries also benefit from distal transfer. The anterior interosseus nerve (just proximal to the pronator quadratus) is transferred to the deep motor branch of the ulnar nerve to restore intrinsic hand function. Another very specific nerve transfer procedure may be used for restoration of pronation. A redundant branch to the FDS may be used. In this case, nerve transfer is quite important as there are limited tendon transfer–based alternatives. The donor nerves are selected based on the proximity to the neuromuscular junction of the target muscle to minimize the reinnervation time, and the repair is done without tension. Nerve grafts are used if necessary; however, a primary repair is preferable, and is the routine. Internal neurolysis allows for dissection of donor and recipient fascicles from the main nerve to enable an end-to-end repair whenever possible. We often perform distal-nerve transfers simultaneous with proximal-nerve graft reconstruction to avoid the formation of a painful proximal neuroma.
Indications and Timing of Nerve Transfers
The surgical indications for nerve transfers continue to evolve as new donor sources for motor and sensory restoration are proposed and evaluated. Because of the excellent results that can be obtained from nerve transfer, we advocate its use in almost any case where regeneration distance and time to reinnervation can be significantly reduced to improve outcome, and to avoid surgery under hostile tissue conditions, especially where other critical structures have already been reconstructed such as vascular injuries. We prefer to use a nerve transfer in managing the following conditions:
Brachial plexus injuries where only very proximal or no nerve is available for grafting
High proximal injuries that require a long distance for regeneration
Avoidance of scarred areas in critical locations with potential for injury to critical structures
Major limb trauma with segmental loss of nerve tissue
As an alternative to nerve grafting when time from injury to reconstruction is prolonged
Partial nerve injuries with a defined functional loss
Spinal cord root avulsion injuries
Nerve injuries wherein the level of injury is uncertain, such as with idiopathic neuritides or radiation trauma and nerve injuries with multiple levels of injury
Recovery of motor function depends on a critical number of motor axons reaching the target muscle and reinnervating muscle fibers within a critical time period. Reinnervation of denervated muscles is generally not possible after 12 to 18 months in adults because of degeneration of the motor end plate. Axonal regeneration occurs at a rate of 1 inch per month or 1 to 1.5 mm/day. The use of distal-nerve transfers can significantly prolong the “window” of opportunity following injury for surgical intervention. A distal-nerve transfer within centimeters of the neuromuscular junction of the target muscle will still have the potential for successful reinnervation even if performed late (8 to 10 months) after the injury.
The use of nerve transfers requires an intimate knowledge of anatomic details of peripheral neuroanatomy. The major nerves of interest derive from the brachial plexus. A few other nerves are mentioned primarily for their use as donor nerves in nerve transfer procedures.
The brachial plexus is the bridge from the spinal cord nerve roots to the terminal nerve branches that innervate the upper extremity ( Fig. 61-1 ). The cervical fifth through eighth (C5–C8) and thoracic first (T1) nerve roots (with variable inclusion of the fourth cervical and second thoracic roots) form the three trunks. The upper (C5 and C6), middle (C7), and lower (C8 and T1) trunks then each divide into an anterior and posterior division. These six divisions then form three cords, with the three posterior divisions becoming the posterior cord, the two anterior upper divisions forming the lateral cord, and the remaining anterior division forming the medial cord. The cords then divide into the terminal nerve branches. The lateral cord goes to the musculocutaneous and median nerves. The posterior cord goes to the axillary and radial nerves. The medial cord goes to the ulnar and median nerves. Note that the median nerve receives components both from the lateral and medial cords.
Specific nerves originate at each segment, with the exception of the divisions, and the pattern of functional deficit will provide clues as to the location and level of injury. Injuries that result in isolated motor or sensory defects are either very proximal or distal. For example, a pure motor injury may indicate a central or motor root lesion or injury to a terminal motor nerve branch such as the anterior interosseous nerve (AIN). A pure sensory nerve injury may involve dorsal root ganglion level injury or injury to a terminal sensory nerve branch such as the lateral antebrachial cutaneous nerve. Mixed patterns of motor and sensory loss indicate an injury in the middle, for example, to the radial nerve. Further examination of function along that nerve helps pinpoint the exact level of injury. For example, if triceps function is maintained but distal wrist and finger extension is lost, the level of radial nerve injury is distal to the takeoff of the nerve branches to the triceps muscle. On the other hand, if a root level injury is suspected, it is important to determine if the injury is very proximal with associated root avulsions. Loss of function of the rhomboids, serratus anterior, or diaphragm hints at a very proximal level of nerve injury. This is because the nerves to these muscles—dorsal scapular, long thoracic, and phrenic, respectively—all originate quite close to the spinal cord. An associated Horner’s syndrome, due to damage to the sympathetic ganglia, also suggests a proximal level of nerve injury. Diagnosis of a proximal or root avulsion injury can significantly change management strategies, because simple repair or interpositional grafting will not be possible.
Other nerves that are relevant for the upper extremity peripheral nerve surgeon include ones used in nerve transfer procedures such as the spinal accessory, intercostal, and phrenic nerves. These are usually not involved in brachial plexus or more distal upper extremity nerve injuries and can serve as relatively expendable donor material. Other nerves of interest are used for interpositional nerve grafting in cases wherein direct end-to-end repair is impossible. Historically, expendable graft material (with the respective length of graft each provides) has been harvested from sensory nerves such as the sural (30–40 cm) and medial (20 cm), and lateral (5–8 cm) antebrachial cutaneous nerves.
History and physical examination are important in the assessment and management of peripheral nerve problems. The extent of injury and any spontaneous recovery should be carefully assessed. Critical information includes the mechanism and time of injury, associated concomitant injuries, and, when considering nerve transfers, the availability of motor nerve branches as potential donor nerves.
As with any functional reconstructive procedure, active and passive range of motion at each joint and function of specific muscles should be assessed and optimized. In chronic injuries, passive range of motion may be limited by joint contracture, which should be treated first by physical or occupational therapy. Motor function may be graded using the standard British Medical Research Council scale. Beginning proximally, strength of shoulder abduction and internal and external rotation are measured. Assess elbow flexion and extension with direct palpation of biceps and brachioradialis musculotendinous units. Forearm pronation, supination, wrist extension, flexion, and radial and ulnar deviation are graded. Hand function, including both intrinsic and extrinsic function, is assessed. An examination of sensation by light touch, evaluation of two-point discrimination, and the ten test also will help to pinpoint the level of injury. The ten test compares sensation in the normal and abnormal areas. As both a normal and abnormal area are touched, the patient is asked to grade the sensation compared to normal or 10 out of 10 sensation. The abnormal area is given a score on a scale from 0, no sensation, to 10, normal sensation. This can be helpful, especially in children as they are often able to at least differentiate between normal and abnormal when the two sides are compared. Putative donor nerves for nerve or tendon transfer should be carefully assessed individually.
Imaging tests can be used to determine associated injuries that may change treatment options as well as demonstrate severity of the primary nerve injury. Plain films of any areas that are suspected of being injured should be performed immediately. For motor vehicle trauma patients these routinely include cervical spine and chest films. For patients with brachial plexus injury, rib fractures on the affected side preclude later use of intercostal nerves for nerve transfer. An elevated hemidiaphragm may indicate a phrenic nerve injury, which would prohibit use of the affected phrenic nerve. An upper extremity angiogram delineates associated vascular injury. Direct assessment of the upper extremity nerve lesion by imaging modalities is still relatively crude. A pseudomeningocele visible on CT myelogram suggests nerve root avulsion. Because the pseudomeningocele may not be visible initially, this test should be done at 3 to 4 weeks postinjury. MRI can also be helpful to show changes such as neuroma formation and serves as a noninvasive method for diagnosis of root avulsion, although CT myelogram remains the gold standard. A normal-appearing CT myelogram or MRI scan does not obviate the need for operative intervention, however.
Nerve conduction studies may be used to assess action potentials across putative lesion sites. If the action potential is preserved, neurolysis alone may be sufficient; lack of a potential is usually fairly definitive evidence that some intervention is required to restore function. Some groups also use sensory evoked potentials for brachial plexus injuries to see if lesions are pre- or postganglionic. This also helps in determining whether proximal exploration and grafting will be of any use at all (preganglionic lesions indicate nerve root avulsion, which is not amenable to this). We rely less on electrodiagnostic tests to identify postganglionic lesions that allow proximal exploration with excision of neuroma and grafting because of our preference for distally based nerve transfer procedures and the often suboptimal outcomes associated with the reconstruction of proximal lesions and long nerve grafts.
Electromyography may show changes indicating recovery in muscles, but the level of recovery may not be clinically sufficient. Therefore, electromyography is most useful when used to serially track function and, especially early on, may justify some watchful waiting. Overall, electromyography is best used as an adjunct to document spontaneous reinnervation as demonstrated by the presence of motor unit potentials. If there is evidence of progressive reinnervation, in a timely fashion after injury, then operative intervention may be unnecessary.
Motor and Sensory Transfers for Isolated Nerve Injuries
Musculocutaneous Nerve—Elbow Flexion
The restoration of elbow flexion is the top reconstructive priority in the patient with brachial plexus injury. Donor muscles used for the restoration of elbow flexion include the pectoralis major, latissimus dorsi, triceps, and the Steindler flexorplasty using the flexor–pronator musculature. Similarly, the nerves that supply these muscles can be used to reinnervate the biceps brachii and brachialis muscles ( Fig. 61-2 ). Historically, the first donor nerves used were either several intercostal nerves with or without interposition nerve grafts or the spinal accessory nerve transferred to the musculocutaneous nerve. In a meta-analysis of the English literature, Merrell et al. found the use of the spinal accessory nerve to be superior to the intercostal nerves for restoration of any function; however, in patients with grade 4 biceps strength, the intercostal nerves were more reliable for transfer to restore strength. In 1993, we described our results with use of medial pectoral nerve branches transferred to the musculocutaneous nerve. This transfer moved the level of the repair more distal than previous reconstructions to allow faster reinnervation, and bypassed the coracobrachialis branch to allow more motor axons to reach the biceps and brachialis muscles. In addition, the lateral antebrachial cutaneous nerve was transposed proximally to neurotize the biceps muscle to redirect motor fibers that had regenerated into the lateral antebrachial cutaneous back to the target muscle.