Peripheral Nerve Injuries

Components of Mixed Spinal Nerves

A typical mixed spinal nerve has three distinct components: motor, sensory, and sympathetic ( Fig. 62.1 ).

Gross Anatomy

Mixed spinal nerves, having left the intervertebral foramina, receive their sympathetic component and promptly branch into anterior and posterior primary rami. The posterior primary rami are directed posteriorly and supply the paraspinal musculature and the skin along the posterior aspect of the trunk, the neck, and the head. The upper three cervical posterior rami are larger than their corresponding anterior rami, supplying relatively large areas of the scalp posteriorly and the musculature around the craniocervical junction. With these exceptions, posterior primary rami are small, and the major part of each spinal nerve continues laterally in an anterior primary ramus to enter a plexus or to become an intercostal nerve.

Anterior primary rami of all the cervical, the first thoracic, and all the lumbosacral nerves join in the formation of plexuses. Alteration of the metameric pattern results from the migration of dermatomes and myotomes into the limb buds. The upper four cervical anterior rami form the cervical plexus, and the lower four cervical and first thoracic anterior rami form the brachial plexus. The first three and a part of the fourth lumbar anterior rami form the lumbar plexus. The sacral anterior rami along with the fifth lumbar and a part of the fourth join to form the lumbosacral plexus. The enlargement and prolongation of the limb bud markedly alter the myotomal pattern, resulting in the union of some myotomes and the division or partial extended migration of others. The fibers of any one mixed spinal nerve may be distributed through several peripheral nerves. By the same token, any one peripheral nerve may contain fibers from several spinal nerves.

The area of skin supplied by the fibers of a single spinal root is called a dermatome. Segmental dermatomal patterns ( Fig. 62.2 ) are well preserved in the thoracic region but not in the limbs. Migration of the limb buds accounts for the displacement of midcervical dermatomes along the lateral aspect of the arm and radial aspect of the forearm and of the lower cervical and upper thoracic dermatomes along the medial aspect of the arm and the ulnar aspect of the forearm. Lumbar and sacral dermatomal alignment along the various aspects of the lower extremity is similarly explained. The line separating the more rostral segmental dermatomes from the more caudal ones is called the axial line and may be followed into the spinal axis.

Microscopic Anatomy

Each nerve fiber, or axon, is a direct extension of a dorsal root ganglion cell (sensory), an anterior horn cell (motor), or a postganglionic sympathetic nerve cell, and it is either myelinated or unmyelinated. Sensory and motor nerves contain unmyelinated and myelinated fibers in a ratio of 4:1 ( Fig. 62.3 ). In the unmyelinated or sparsely myelinated fibers, several axons are wrapped by a single Schwann cell. In the more heavily myelinated fibers, the Schwann cell by rotation forms a multilaminated structure that encloses a myelin sheath around a single axon. The segment of myelinated nerve fiber enclosed by a single Schwann cell is referred to as an internode and varies in length from 0.1 to 1.8 mm, with the more heavily myelinated fibers having the longer internodes. The point at which one Schwann cell ends and the next begins is relatively sparse in myelin and is called the nodal gap, or node of Ranvier ( Fig. 62.4 ). The axon with its Schwann cell and myelin sheath is surrounded by a veil of delicate fibrous tissue called the endoneurium. Seen longitudinally, the endoneurium forms a tube encircling individually the Schwann cell sheaths that cluster together to form a fascicle (or funicle as termed by Sunderland). Each fascicle or separate group of sheathed axons is surrounded by a denser layer of perineurium. The entire group of fascicles with their surrounding perineurium is encased as a mixed spinal or peripheral nerve in a denser epineurium. The blood supply to the peripheral nerve enters through the mesoneurium, which is loose connective tissue extending from the epineurium to the surrounding tissues. There is an extrinsic (segmental) and an intrinsic (longitudinal) blood supply to each nerve. The intrinsic blood supply that runs longitudinally within the epineurium, perineurium, and endoneurium is fairly extensive and allows surgical mobilization without complete devascularization over variable lengths of nerves.

Motor

When a peripheral nerve is severed at a given level, all motor function of the nerve distal to that level is abolished. All muscles supplied by branches of the nerve distal to that level are paralyzed and become atonic. Significant changes on electromyography (EMG) are not apparent for 8 to 14 days, at which time transient fibrillation potentials on needle insertion may become apparent. Spontaneous fibrillations may become evident after 2 to 4 weeks, coinciding with the onset of atrophic change within the muscle fibers. Atrophy of muscle bulk progresses rapidly to 50% to 70% at the end of about 2 months. Atrophy continues at a much slower rate, and the connective tissue component of the muscles increases. Striations and motor endplate configurations are retained for longer than 12 months, whereas the empty endoneurial tubes shrink to about one third their normal diameter. Complete disruption and replacement of muscle fibers may not become complete until after 3 years.

Several methods are used to evaluate motor return after peripheral nerve injuries. They involve assessment of muscle strength against gravity and against graded resistance. The use of pinch meters and grip meters and evaluation of endurance, speed of movement, and individual muscle function helps to document the progress of motor return. The British Medical Research Council established the following system for assessing the return of muscle function after peripheral nerve injuries: M0, no contraction has returned; M1, perceptible contraction in proximal muscles has returned; M2, perceptible contraction in proximal and distal muscles has returned; M3, all important muscles act against resistance; M4, all synergistic and independent movements are possible; M5, recovery is complete ( Table 62.2 ).

Sensory

Sensory loss usually follows a definite anatomic pattern, although the factor of overlap from adjacent nerves may confuse inexperienced surgeons. After severance of a peripheral nerve, only a small area of complete sensory loss is found. This area is supplied exclusively by the severed nerve and is called the autonomous zone or isolated zone of supply for that nerve. A larger area of tactile and thermal anesthesia is readily delineated and corresponds more closely to the gross anatomic distribution of the nerve ( Fig. 62.8 ); this larger area is known as the intermediate zone. When a nerve is intact, and the adjacent nerves are blocked or sectioned, an area of sensibility exceeds the gross anatomic distribution of the nerve; this area is known as the maximal zone.

It has long been recognized that the autonomous zone becomes smaller during the first few days or weeks after injury, long before regeneration is possible. Livingston suggested that this is caused by ingrowth of adjacent nerves, but resumption of or increase in function in anastomotic branches from adjacent nerves is a more plausible explanation. This decrease in the area of sensory loss might be interpreted by an inexperienced surgeon as evidence of regeneration or of incomplete injury and might be responsible for needless delay in exploration of the nerve.

In injury to the median and ulnar nerves, one study found that pinprick was the first perception to return, followed by 30 cycles/s vibratory stimulus, and then moving touch. The perception of constant touch and the perception of a 256 cycles/s vibratory stimulus were the last to return ( Table 62.3 ). These investigators inferred that the early return of pain perception resulted from the faster regeneration of the small-diameter pain fibers. The larger-diameter touch fibers regenerated more slowly. The return of moving touch perception, mediated by quickly adapting fibers and Pacinian corpuscles, before the return of constant touch, mediated by slowly adapting fibers and the Merkel discs, was explained by differential maturation of the respective receptors, rather than by the diameter of the fibers alone. The system of evaluating by moving touch, constant touch, vibratory stimulus, pinprick, and the Weber two-point discrimination was proposed as a method for screening patients to determine specific exercises for reeducation of constant-touch perception. The evaluation of sensory return after peripheral nerve injuries is important regardless of the site of the injury. This is especially true in the upper extremity, where sensibility in the hand is extremely important.

The clinical evaluation of sensory return also is done using other methods, such as pinprick appreciation and von Frey hairs. The British Medical Research Council established the following six-level grading scale for sensory return: S0, absence of sensibility in the autonomous area; S1, recovery of deep cutaneous pain within the autonomous area; S2, return of some superficial cutaneous pain and tactile sensibility within the autonomous area of the nerve; S3, return of superficial cutaneous pain and tactile sensibility throughout the autonomous area with disappearance of overreaction; S3+, some recovery of two-point discrimination within the autonomous area; S4, complete recovery ( Table 62.4 ).

Two-point discrimination has been shown to directly correlate with return of hand function and object identification. The pick-up test (a timed test to measure fine and manual dexterity) and the triketohydrindene hydrate (ninhydrin) printing test also have been shown to be of use.

Reflex

Complete severance of a peripheral nerve abolishes all reflex activity transmitted by that nerve. This is true in severance of the afferent or the efferent arc. Commonly, however, reflex activity is abolished in partial nerve injuries when neither arc is completely interrupted and is not a reliable guide to the severity of injury.

Autonomic

Interruption of a peripheral nerve is followed by loss of sweating and of pilomotor response and by vasomotor paralysis in the autonomous zone. The area of anhidrosis usually corresponds to, but may be slightly larger than, the sensory deficit. This area may be outlined easily by the starch-iodine test, by the ninhydrin printing test popularized by Aschan and Moberg, or by instruments for determining skin resistance (Richter dermometer). Another objective test described by O’Riain and by Leukens is the wrinkle test. When normal skin is immersed in water for a time, wrinkling occurs. Denervated skin does not wrinkle under these circumstances. As reinnervation occurs, wrinkling of the skin returns. If the injury is incomplete, and especially if it is associated with causalgia, sweating may be excessive and may involve areas beyond the intermediate zone of the nerve. Vasodilation occurs in complete lesions, and the area affected is at first warmer and pinker than the rest of the limb. After 2 to 3 weeks, however, the affected area becomes colder than the adjacent normal areas and the skin may be pale, cyanotic, or mottled in an area often extending beyond the maximal zone of the injured nerve. Trophic changes occur commonly and are most evident in the hands and feet. The skin becomes thin and glistening and, when subjected to trauma that ordinarily does little harm, breaks down to form ulcers that heal slowly. The fingernails become distorted, are often ridged or brittle, and may be lost entirely.

Osteoporosis often follows peripheral nerve injuries. It is more likely to be pronounced in incomplete lesions associated with pain. Incomplete lesions of the median nerve seem to be associated more often with osteoporosis, with changes occurring in the distal phalanges of the thumb and index and long fingers. Partial ankylosis from fibrosis of the periarticular structures also may develop. These changes are similar to atrophy of disuse but are much more severe.

Complex Regional Pain Syndrome (Reflex Sympathetic Dystrophy)

Complex regional pain syndrome (CRPS) represents autonomic and pain transmission dysregulation, resulting in peripheral sensitization with allodynia, dysesthesia, hyperpathia, and a reduced tolerance for pain when using the affected area for basic function. The condition occurs most commonly after a traumatic injury or iatrogenic insult. Classification of CRPS is based on the structures injured. The International Association for the Study of Pain delineated two main categories of the syndrome, replacing traditional terms, and through efforts by their Taxonomy Committee ( iasp-pain.org ), they continue to update descriptions and terms. Although the attempt to simplify CRPS into types I and II is attractive, considerable overlap in pathology exists. CRPS I (formally reflex sympathetic dystrophy , RSD) theoretically represents patients who have had a musculoskeletal injury without a defined neural injury. CRPS II (causalgia) includes patients who fulfill the same criteria but who have evidence of a neural injury. In addition, there have been further efforts to define sympathetic-mediated and nonsympathetic-mediated varieties, with temporal transitions into “warm” and “cold” subtypes.

Diagnostic Tests

Electrodiagnostic Studies

The best and most accessible correlative electrophysiologic confirmations of a peripheral neural injury are nerve conduction and electromyographic mapping assessments. The surgeon must have specific objectives when ordering these tests to obtain the most useful information for clinical management. The timeline in ordering the studies is also important because the changes after injury and recovery follow a well-described pattern. The presence, location, severity, and possibly the prognosis of the neural insult can be determined from these studies, and information regarding the recovery pattern can be obtained when the study is done sequentially over time. Alternative electrophysiologic uses include dynamic electromyographic assessment when considering optimal muscle transfer strategies, before tenotomy, or botulinum toxin injections in central and peripheral neuropathic conditions. Electrical stimulation can be used for optimal nerve localization when considering blocks or ablation procedures. Generally, both nerve conduction velocity studies and EMG are ordered for routine neural injury assessments because the information gained is complementary. Although full neuropathic changes are not observed in these studies for 2 or 3 weeks after injury, there may be instances in which early baseline studies should be done.

Nerve Conduction Velocity

Standard nerve conduction techniques include orthodromic motor and antidromic-orthodromic sensory studies and retrograde studies (e.g., F wave study). F wave studies are especially useful for investigating peripheral nerve injuries that are more proximal and less accessible through other techniques. The suspected location of neural compromise is identified, and a protocol to electrically stimulate proximally, distally, and across the segment is formulated. Depending on whether it is an orthodromic or antidromic study, the evoked potential will be recorded at some defined point proximal and distal to the injury with a surface or needle electrode ( Fig. 62.9 , Segment A ).

After a severe traumatic neural insult and Wallerian degeneration, there is a progressive structural degradation and neurotransmitter compromise expressed by alteration in nerve conduction and evoked motor and sensory configuration. Immediately after injury, conduction proximal and distal to the insult usually elicits a normal response, although stimulation across the injured segment may vary, depending on the presence of axonal or myelin injury. As Wallerian degeneration ensues (within 5 to 10 days), there is a progressive reduction in the amplitude and alteration in the configuration of the evoked potentials ( Fig. 62.9 , Segment B ). If the insult produces only a temporary physiologic block (e.g., neurapraxia), conductivity distal to the lesion remains preserved even after 10 days and a more favorable prognosis can be expected. Evoked sensory amplitude assessments and comparisons also can assist in delineating further pathology.

With a more severe injury, not only is a conduction block across the segment present but a progressive decline in amplitude is noted in evoked potentials when stimulating distal to the injury; sometimes there is complete absence of a response (e.g., axonotmesis). Eventually, electromyographic changes evolve. Over a period of months, repeat studies may be performed to follow neural recovery patterns depending on the case.

Electromyography

Manual muscle testing is routine with any musculoskeletal examination, but it is not sensitive for picking up more subtle neuropathic pathology. Myotomal sampling of the involved extremity with needle pick-up electrodes (e.g., monopolar, concentric, and single fiber) yields information regarding neuropathic injury and pathology. It can distinguish a recent injury from a chronic condition that predated the injury (e.g., workers’ compensation or litigation). The basic monopolar needle electrode samples approximately eight muscle fibers, and by assessing different sites, fair representation of specifically innervated myotomal groups is possible. The muscle initially is observed at rest (insertional activity, approximately 200 ms) and subsequently during volitional muscle recruitment. During the initial postinjury phase, needle sampling should be normal unless there has been a prior injury ( Fig. 62.10 ). Recruitment at this point may vary depending on injury pattern and effort. At 10 to 14 days after neural injury, abnormal spontaneous rest potentials evolve (positive sharp waves) appearing in denervated myotomes where axonal injury has occurred ( Fig. 62.11A ). Between 14 and 18 days, fibrillations appear ( Fig. 62.11B ). Voluntary motor unit potentials, if present, may have attenuated amplitudes reflecting axonal compromise. Abnormal denervation patterns can be correlated with established intraneural topographic reference guides to assist in a clear anatomic mapping of the injury. Abnormal spontaneous rest potentials may last indefinitely until the muscle has become reinnervated or fibrotic.

At approximately 3 months after injury, some peripheral neural sprouting occurs and the motor unit potential amplitude progressively increases; this is preceded at times by polyphasic configuration potentials. Between 2 and 6 months after injury, larger than normal appearing potentials are established and remain so until the reinnervation is completed, at which time the motor unit potential configuration returns to a more normal-appearing pattern. Some surgeons monitor denervated myotomes over months (e.g., 3 months) before exploration, depending on the nature of injury and clinical presentation.

Age

Age undoubtedly influences the rate and degree of nerve regeneration. All other factors being equal, neurorrhaphies are more successful in children than in adults and are more likely to fail in elderly patients; why this is true has not been completely explained, but it may relate to the potential for central adaptation to the peripheral nerve injury. We do not know precisely what results can be expected in either of the extremes of age because practically all significant studies have dealt with military personnel whose average age was 18 to 30 years. A close correlation has been noted between age and two-point discrimination obtained after median and ulnar nerve repairs (30 mm at 20 to 40 years; 15 mm at 11 to 20 years; 10 mm at <10 years). After digital nerve repair, however, the final two-point discrimination was not as closely related to age. Another study found that a higher percentage of patients younger than 20 years at the time of repair had two-point discrimination of less than 6 mm than did patients older than 20 years.

Gap Between Nerve Ends

The nature of the injury is the most important factor in determining the defect remaining between the nerve ends after any neuromas and gliomas are resected. When a sharp instrument, such as a razor or knife, severs a nerve, damage is slight proximally and distally, and although the nerve ends inevitably do retract, the gap can usually be easily overcome. Conversely, when a high-velocity missile severs a nerve, proximal and distal nerve damage is extensive. Ultimately, both ends must be widely resected to expose normal funiculi, producing a larger gap. The gap is increased farther if part of the nerve is carried away by a missile, as in shrapnel injuries. Methods of closing troublesome gaps include (1) nerve mobilization, (2) nerve transposition, (3) joint flexion, (4) nerve grafts, and (5) bone shortening. The greater the defect, the more dissimilar the funicular pattern of the two ends because of the constantly changing arrangement of fibers within the nerve as it progresses distally. This is particularly important in the more proximal portion of peripheral nerves. Agreement is widespread that excessive tension on a neurorrhaphy harms nerve regeneration. Nerve grafting is advised if, after the nerve is mobilized, the gap cannot be closed by flexing the main joint of the limb 90 degrees. The observed upper limit of a gap beyond which results deteriorate is approximately 2.5 cm. The observations of Kirklin, Murphey, and Berkson in 1949 that recovery is slightly better when the gap is relatively small remain valid.

Delay Between Time of Injury And Repair

Delay of neurorrhaphy affects motor recovery more profoundly than sensory recovery, most likely because of the survival time of denervated striated muscle. There is significant loss of motor endplates and increased muscle fibrosis by 18 months after denervation; therefore nerve repair needs to be performed early enough to allow reinnervation of muscle before this occurs. Experimental studies have shown better axonal survival with early nerve repair.

As a rule of thumb, Omer suggested that about 1% of recoverable nerve function is lost for each week of delay after 3 weeks postinjury. The influence of delay on sensory return is unclear; in the Veterans Administration study, little influence could be found and useful sensation returned in a few patients when suture was performed 2 years after injury. The critical limit of delay beyond which sensation does not return is unknown.

Our practice is to perform neurorrhaphies in clean, sharp wounds immediately or during the first 3 to 7 days. In the presence of extensive soft-tissue contusion, laceration, crushing, or contamination in which the proximal and distal extent of the nerve injury is impossible to delineate, a delay of 3 to 6 weeks is preferred.

Level of Injury

The more proximal the injury, the more incomplete the overall return of motor and sensory function, especially in the more distal structures. Conditions are more favorable for recovery in the more proximal muscles because (1) the neurons that innervate the distal portions of the limb are more severely affected by retrograde changes after proximal injury, (2) a greater proportion of the cross-sectional area of the nerve trunk is occupied by fibers to the proximal muscles, and (3) the potential for disorientation of regrowing axons and for axon loss during regeneration is greater for the distal muscles than for the muscles more proximally situated after a proximal injury. Except for parts of the brachial plexus, useful function at times returns regardless of the level of injury if the critical limit of delay has not passed.

Condition of Nerve Ends

The condition of the nerve ends at the time of neurorrhaphy is important. Meticulous handling of the nerve ends, asepsis, care with nerve mobilization, preservation of neural blood supply, avoidance of tension, and provision of a suitable bed with minimal scar all exert favorable influences on nerve regeneration. Distal stump shrinkage has been found to be maximal at about 4 months, leaving the distal fascicular cross-sectional area diminished to 30% to 40% of normal size. Intraneural plexus formation and fascicular dispersal make accurate fascicular alignment and appropriate axonal regeneration more difficult. A neurorrhaphy with a satisfactory external appearance is no guarantee of optimal internal fascicular alignment. Fascicular malalignment is a common finding. It is generally agreed that the nerve ends should be prepared in such a way that a satisfactory fascicular pattern is apparent in the proximal and distal stumps. No scar, foreign material, or necrotic tissue should be allowed to remain around the ends to interfere with axonal regeneration. Sometimes resection of the nerve ends so that satisfactory fasciculi are exposed leaves a gap that cannot be closed by end-to-end repair. As noted previously, clinical and experimental evidence indicates that excessive tension on the neurorrhaphy at the time of repair and when an acutely flexed limb is mobilized later causes excessive intraneural fibrosis. These findings and the promising results achieved after the interfascicular nerve grafting technique advocated by Millesi and by Millesi, Meissl, and Berger suggest that such a technique is preferable to repair of nerves under too much tension or with limbs in acutely flexed or awkward positions.

Indications

In the presence of a traumatic peripheral nerve deficit, exploration of the nerve is indicated as follows:

• 1.

  When a sharp injury has obviously divided a nerve, early exploration is indicated for diagnostic, therapeutic, and prognostic purposes. Neurorrhaphy can be done at the time of exploration or can be delayed.

• 2.

  When abrading, avulsing, or blasting wounds have rendered the condition of the nerve unknown, exploration is required for identification of the nerve injury and for marking the ends of the nerve with sutures for later repair.

• 3.

  When a nerve deficit follows blunt or closed trauma and no clinical or electrical evidence of regeneration has occurred after an appropriate time, exploration of the nerve is indicated. This also is true when a nerve deficit complicates a closed fracture. In this instance, it has been our practice to observe the patient for evidence of nerve regeneration for an appropriate time, depending on the nerve and its level of muscle innervation. Then if regeneration has not occurred, we favor exploration. In situations in which a nerve has been intact before closed reduction and casting of a fracture, but a significant deficit is found immediately after, we explore the nerve as soon as feasible.

• 4.

  When a nerve deficit follows a penetrating wound, such as that caused by a low-velocity gunshot, the part is observed for evidence of nerve regeneration for an appropriate time. If there is no evidence of regeneration, exploration is indicated.

Conversely, delay in exploration of a nerve injury is indicated if progressive regeneration is evidenced by improvement in sensation, motor power, and electrodiagnostic tests and by progression of the Tinel sign.

Time of Surgery

It has been the time-honored policy to advise primary suture when possible. This recommendation is logical when one considers what happens to the distal end of the nerve, motor endplates, sensory nerve ends, muscles, joints, and other tissues of the denervated extremity. The controversy concerning whether primary or secondary nerve repair is better is unresolved. Primary repair done in the first 6 to 8 hours or delayed primary repair done in the first 7 to 18 days is appropriate when the injury is caused by a sharp object, the wound is clean, and there are no other major complicating injuries. Ideally, such repairs should be performed by an experienced surgeon in an institution where adequate equipment and personnel are available. The development of magnification devices, new instruments, and new techniques and the modification of a variety of small instruments for use in nerve surgery have improved the technique of early repair. Primary repair should shorten the time of denervation of the end organs, and fascicular alignment should be improved because minimal excision of the nerve ends is required. Regarding war wounds, however, primary sutures have compared unfavorably with early secondary suture.

When the diagnosis of division of a peripheral nerve has been made, if conditions are suitable and repair is indicated, one should not delay repair in anticipation of spontaneous regeneration. Only if the patient’s life or limb is seriously endangered should the operation be long postponed. A fracture is not a contraindication for operation. Operation before the fracture becomes united may be advantageous for two reasons: (1) if bone shortening is necessary, resection of an ununited or partially united fracture is a much less formidable procedure than resection of a fully united bone; and (2) restriction of joint motion is minimal if the nerve is repaired soon after the injury; later, motion would be more limited, perhaps so severely as to prevent flexing the joint enough to overcome a gap between the nerve ends.

Instruments and Equipment

A nerve stimulator should be available for all peripheral nerve procedures; many satisfactory permanent and disposable ones are available commercially. A stimulator is indispensable in investigating partially severed nerves and neuromas in continuity and in locating and preserving nerve branches given off proximal to or at the lesion that are still functioning but are encased in scar tissue. Intraoperative recording of somatosensory evoked potentials and nerve action potentials is useful in surgical planning and assessing nerve lesions. These techniques require sensitive and sophisticated recording and monitoring equipment and trained technicians. (For details of these monitoring techniques, the reader is referred to the references at the end of this chapter.)

Despite the technical difficulties involved in these methods, we have found intraoperative recording to be helpful when evaluating partial nerve lesions and neuromas in continuity. Instruments for handling and dissecting delicate tissues always are essential. Nerve surgery in the extremities also is made easier by the use of a pneumatic tourniquet, suction apparatus, and bipolar electrocautery. Gelfoam and thrombin are useful for controlling the bleeding from the cut ends of nerves. For suture material, we prefer 8-0, 9-0, and 10-0 monofilament nylon. The tensile strength, easy handling qualities, and minimal tissue reaction of nylon make it the most desirable suture material now available for neurorrhaphy. In our experience, most epineurial repairs are best done with 8-0 or 9-0 nylon. For perineurial or epiperineurial repair, 9-0 or 10-0 monofilament nylon is preferable.

Anesthesia

Peripheral nerve operations can be done with the patient under general, regional, or local anesthesia for the upper extremities or general, spinal, or local anesthesia for the lower extremities. Local anesthesia has the advantage of allowing evaluation of the passage of sensory impulses through the injured nerve. If evaluation is to be accurate, however, little if any anesthetic agent should be injected around the nerve, and, consequently, the procedure is painful. There is always the possibility that the agent would infiltrate the tissues around the nerve and interfere with motor response to stimulation. As a rule, we prefer general anesthesia for surgery in the upper extremities and neck and general or spinal anesthesia for surgery in the lower extremities.

Preparation and Draping

Before preparing and draping, the correct side and site are identified and the site is marked with an indelible surgical marking pen. Because the exact length of an incision can rarely be predicted, it is mandatory that the entire extremity and its environs be prepared. For an operation on the upper extremity, the axilla, shoulder, neck, and chest should be included in the field of preparation; for an operation on the lower extremity, the buttock and the area up to the iliac crest posteriorly should be included. In operative procedures involving the distal portions of nerves only, such as below the elbow or knee, a well-padded pneumatic tourniquet placed above the elbow or knee is used, limiting the sterile field. A sterile tourniquet also can be helpful for more proximal lesions.

After preparation of the entire field, the proposed incision is marked on the extremity and is crosshatched with washable ink before any of the landmarks are covered. It is a good policy to mark the incision along the course of the nerve in the entire prepared area. The extremity is encased in a sterile stockinette so that it can be moved freely over the sterile drapes. If it is desirable to watch the movement of the muscles in the hand when the nerve is stimulated, the hand can be left exposed and bare.

Endoneurolysis (Internal Neurolysis)

When an endoneurial exploration is undertaken, it should be borne in mind that neurolysis or partial or complete neurorrhaphy may be necessary, and one should preserve intact as much of the epineurium and normal nerve as possible. The epineurium is incised longitudinally proximal to the lesion, beginning not more than 0.5 cm from the level of gross changes in the nerve as determined by palpation. The incision is not extended more proximal to this point unless necessary because the epineurium may become frayed; and if neurorrhaphy becomes necessary, more of the nerve may have to be sacrificed. For the same reason, the distal end of the incision is limited. The flaps of epineurium on each side may be retracted laterally by nylon sutures and are undermined widely. The funiculi are separated if possible with a pointed or diamond-bladed knife, using sharp or blunt dissection as necessary. Spring-loaded microscissors also are helpful in this dissection. The surgeon constantly should be aware of the possibility of plexus formation between fascicles and protect these. Distinguishing between intraneural fibrosis and plexus formation is extremely difficult. If most of the fasciculi are intact and can be separated and traced through the neuroma, nothing further should be done. If stimulation fails to elicit a response, and few if any intact fasciculi can be found, resection of the neuroma and neurorrhaphy are probably indicated. Use of magnifying loupes or the operating microscope is essential when performing intraneural dissection to avoid injury to intact nerve tissue (see Chapter 63 ).

Partial Neurorrhaphy

Partial severance of the larger nerves, such as the sciatic nerve and the cords and trunks of the brachial plexus, is common. In such an injury, partial neurorrhaphy is best. It is occasionally necessary and justifiable in smaller nerves but is never quite as satisfactory technically as is complete neurorrhaphy. The decision to perform partial neurorrhaphy is likewise often difficult. The decision should be made only after the most careful investigation of the lesion. If one half of the nerve, especially a large one, is disrupted, partial neurorrhaphy is advisable. If the motor response to stimulation is good, however, it would be unwise in some nerves, such as the peroneal or ulnar, to risk injury of good motor funiculi in an attempt to restore sensation to a small area on the dorsum of the foot or to the little finger. If most of the fascicles in smaller nerves are severed, and if stimulation cannot show important function in the few that remain, complete neurorrhaphy probably is better. Suture of a few fascicles usually is impractical.

When the decision has been made to perform partial neurorrhaphy ( Fig. 62.12 ), the incision is extended longitudinally in the epineurium proximally and distally several centimeters, as necessary. The intact funiculi are dissected out for the same distance. The ends of the injured part of the nerve are resected to normal tissue. At the cut ends, an end-to-end neurorrhaphy is performed. If the epineurium is inadequate for placement of epineurial sutures, epiperineurial or perineurial (fascicular) sutures suffice. The proximal and distal dissection should be extensive enough to prevent kinking of the loop of intact nerve.

Neurorrhaphy and Nerve Grafting

When a nerve has been completely severed, and when conditions as already outlined are appropriate, neurorrhaphy after sufficient resection of the proximal and distal ends of the nerve is indicated. Sometimes a considerable gap or defect (actual loss of nerve tissue) remains after excision of any glioma and neuroma, and selecting a method for overcoming the gap is difficult. Extension of the incision proximally and distally can be helpful in permitting adequate dissection for closure of the gap. In general, direct neurorrhaphy may be possible with fairly large gaps in the median and ulnar nerves near the wrist and elbow after mobilization of proximal and distal segments, whereas gaps of 2 to 3 cm in the brachial plexus and radial, sciatic, and peroneal nerves and the median nerve at the midforearm level may require nerve grafting. Regardless of the technique used, there is general agreement that nerve repair under excessive tension is detrimental to satisfactory regeneration. It generally is recommended that if a single 8-0 nylon epineurial suture can maintain approximation of the nerve ends, excessive tension is not present.

Techniques of Neurorrhaphy

Fibrin clot, micropore tape, collagen tubulization techniques, adhesives, and many varieties of sutures and suture techniques have been proposed for neurorrhaphy. Neurorrhaphy by suture with nonreactive and nonabsorbable materials, such as stainless steel and monofilament nylon, has the widest application and acceptance. Magnification, appropriate small instruments, and meticulous technique are essential. Experimental evidence is conflicting concerning the relative merits of epineurial and perineurial (fascicular) neurorrhaphy techniques. Clinical evidence to support the use of one technique over the other is meager and inconclusive. The technique selected by the individual surgeon depends on training and experience. Proponents of repair supplementation with autologous fibrin glue, or other commercially available “nerve glues,” cite less tendency for gapping at the repair site, fewer sutures required for the repair, and a possible barrier to invading scar tissue as advantages. None has been shown to increase the strength of the repair, and only one has been shown not to block axonal regeneration when interposed between nerve ends. Our preference is epiperineurial repair at the periphery of the nerve combined with perineurial (fascicular) neurorrhaphy for large fascicles within the nerve. Sunderland pointed out that funicular (fascicular) repair cannot be done accurately in every instance because (1) funicular patterns at nerve ends match exactly only after clean transection, (2) the numbers of funiculi at nerve ends may not correspond, and (3) any discrepancies in funiculi within the nerve would require excessive intraneural suture material. He suggested that funicular repair might be practical when (1) funicular groups are large enough to take sutures that maintain funicular apposition, (2) nerve ends show a funicular pattern that would predispose to wasteful regeneration of axons if epineurial repair were done, and (3) each funicular group is composed of nerve fibers to a particular branch occupying a constant position at the nerve ends. The last arrangement can be seen in the median and ulnar nerves at and above the wrist and the radial nerve at and just proximal to the elbow. Sunderland recommended suturing groups of funiculi in such situations.