Hand Surgery
1st Edition

46
Chronic Nerve Injuries and Treatment of Neuromas
Lee E. Edstrom
Ercan Karacaoglu
Chronic nerve injuries in the upper extremity can result in severe disability. Injuries to peripheral nerves are common in all forms of upper extremity trauma, including fractures, dislocations and lacerations, avulsion, and crush or amputation injuries. Patients are usually young adults in the various activities of life, so the effect of these injuries on the nation’s economy is considerable.
One of the most devastating sequelae of upper extremity injuries is the painful neuroma (Fig. 1). A simple painful neuroma may impair function of the whole extremity. Most occur in the extremity and are usually associated with direct nerve injury. Although neuromas may develop in all nerves, only the ones containing sensory fibers become painful.
In the past three decades, microsurgical techniques and instruments in nerve repair have resulted in improved outcomes for nerve injuries. However, unsatisfactory results are all too frequent, and much more improvement is needed. It is currently stated that we are in the transition from a mechanical to a molecular view of the events of repair and regeneration. Pharmacologic and genetic manipulations to change the process of degeneration and regeneration will presumably lead to fundamental improvements in the treatment of chronic nerve injuries.
NORMAL AND PATHOLOGIC SURGICAL ANATOMY
The constituent fibers of all but the smallest peripheral nerves are arranged in bundles, or fasciculi, and three connective tissue sheaths are recognized. The entire nerve is surrounded by the epineurium. This is composed of ordinary connective tissue (circumferential epineurium), and it also fills the spaces between the fasciculi, which contain multiple vascular channels (intraneural epineurium). The sheath that encloses each small fascicle in a nerve consists of several layers of flattened cells, collectively known as the perineurium. Within the perineurium, individual nerve fibers have a delicate covering of connective tissue that constitutes the endoneurium, or sheath of Henle (1,2).
Perineurium has a well-differentiated ultrastructure that must be regarded as a boundary between two environments: endoneural space and the connective tissue. The perineurium consists of several layers of flattened cells and collagen fibers arranged longitudinally and sometimes circumferentially. Their diameter varies between 400 and 800 Å, and together their total thickness is 5 μ (3).
The importance of perineurium in surgical procedures is always stressed, as it has a fragile structure and damage to it results in an axonic explosion (4,5).
When the nerve is cut, its fascicular contents with its gelatinous intrafascicular tissue ooze out, as if under pressure, from the perineural sheath, surface tension being responsible for the characteristic translucent club-shaped appearance of the fascicular contents pouting out of the perineurium (3).
It has been shown by Sunderland that peripheral nerves exhibit branching and fascicular plexus formation (6). Many surgeons assumed this pattern occurred at all levels and that it followed a tortuous course of plexus formation until the fibers finally organized themselves into specific groups distally. It was shown, however, by Jabaley et al. that fibers destined for a specific territory organized themselves into distinct groups proximally within the nerve (7). Even though their position within the nerve might change, they could be traced for considerable distances proximally.
Lundborg and Branemark systematically demonstrated the vascular supply at peripheral nerves. Their findings with regard to microvascularization of the peripheral nerves revealed that the nerve has a dual vascular system: an extrinsic (perifascicular) system and an intrinsic (intrafascicular) system, which communicate freely (8,9).
Clinically, the nerve right after it is transected has the appearance of axoplasm bulging from the cut ends of the axon. At 2 or 3 weeks after injury, however, nerve has a thickened epineurium (10).
FIGURE 1. A traumatic peripheral neuroma.
P.848

After nerve injury, disorganized nerve fibers grow at the site of injury if distal connections are not made (11). Scar is deposited in concert with this axonal sprouting and results in the formation of a mass or nodule that represents the neuroma (Fig. 2) (12).
Histopathologic examination of human neuroma tissue demonstrates regenerating nerve fibers embedded in collagenous scar. The regenerating fibers are noted to be randomly oriented. Transverse sections show longitudinal and transverse fibers surrounded by dense connective tissue; this outcome is a direct result of classic regenerative mechanisms of injured peripheral nerve (12).
PATHOPHYSIOLOGY
When considering the complex microanatomy of a peripheral nerve trunk, it is obvious that any injury (e.g., laceration, crush, stretch, contusion, radiation) may affect various intraneural tissue components in different ways. Axons are well protected by connective issue and may therefore not be damaged by direct injuries. A moderate trauma to a nerve trunk may induce a tissue reaction in the most superficial layers of the nerve, the epineurium, which results in microvascular stasis, edema, and long-term intraneural fibrosis. Such an inflammatory response may have secondary effects on the nerve fibers in a complex manner. Influence of trauma on intraneuronal transport of different constituents, the “axonal transport,” may have consequences on the structure and the function of parts of the neuron far proximal and distal from the compression site, including the nerve cell body (13).
FIGURE 2. Hematoxylin-eosin stain section of peripheral neuroma demonstrating disorganized neural, fibrous, and vascular elements.
Definition and Classification
In the spectrum of these different injury mechanisms of nerves, a neuroma is the inevitable result of endoneurial disruption as regenerating axons escape previously defined anatomic pathways and enter epineural and extraepineural tissue layers (14,15).
Nerve injuries may be complete or incomplete. Seddon’s classification divided the nerve injuries into three types: (a) neuropraxia, (b) axonotmesis, and (c) neurotmesis (16).
Sunderland contributed a further subdivision of nerve injuries and listed five grades according to their severity: grade I, loss of conduction in the axons; grade II, loss of continuity of the axons without affecting the endoneurium; grade III, loss of continuity of the nerve fibers (endoneurium affected); grade IV, loss of continuity of the fascicles (perineurium affected); and grade V, loss of the entire nerve (Table 1) (17).
TABLE 1. CLASSIFICATION SYSTEMS FOR NERVE INJURIES
Seddon classification Sunderland classification Pathology
Neuropraxia Grade I Myelin injury or ischemia
Axonotmesis Axons disrupted, variable stromal disruption
Grade II Axons disrupted, endoneurial tubes intact, perineurium intact, epineurium intact
Grade III Axons disrupted, endoneurial tubes disrupted, perineurium intact, epineurium intact
Grade IV Axons disrupted, endoneurial tubes disrupted, perineurium disrupted, epineurium intact
Neurotmesis Grade V Axons disrupted, endoneurial tubes disrupted, perineurium intact, epineurium disrupted
TABLE 2. CLASSIFICATION OF THE NEUROMAS BASED ON MICROSCOPIC INJURY PATTERN
Neuroma-in-continuity
   Spindle neuroma
   Lateral neuroma
   Neuroma after nerve repair
Neuroma in completely severed nerve
Amputation stump neuroma
P.849

These two classifications can be compared for purposes of description. For example, proximal pseudoneuromas as seen in nerve compression problems such as the carpal tunnel syndrome can range in severity from grade I to grade III in Sunderland’s classification and may represent either neuropraxia or axonotmesis in Seddon’s classification. True neuromas form after complete nerve disruption and are therefore classified under neurotmesis (Seddon) or as grade V lesions (Sunderland) (18).
To simplify the management of neuromas, two different types of classification have been devised. First, anatomic classification is based on the location of the neuroma in the human body, such as Morton’s neuroma (foot), finger amputation neuroma, acoustic neuroma, and so forth. An anatomic classification of neuromas is not necessarily needed but may be helpful in explaining patient management.
Second, gross morphology is used by some authors to clarify neuromas. However, recent preference is to classify neuromas based on microscopic injury patterns that relate well to functional status. The latter is also modified by Herndon (Table 2) (11,12,13,14,15,16,17,18,19 and 20).
Neuroma-in-Continuity
Neuroma-in-continuity is a Sunderland grade 4 or 6 injury, implying loss of axonal continuity to varying degrees and often involving several different fascicles simultaneously. Two subtypes are described: spindle neuromas and lateral neuromas (11).
Spindle Neuromas
Spindle neuromas are lesions in which perineural sheath is intact. They are swellings in an intact nerve secondary to chronic irritation, friction, or pressure. Sometimes, they are termed pseudoneuroma. Histologically, the bulbous area contains increased connective tissue. The fibrous tissue proliferates with continuous irritation, and finally the enlargement becomes a large collagenized mass. In time, fibrotic tissue replaces the nerve fibers and vessels (11,17).
Lateral Neuromas
Lateral neuromas are lesions in which the perineurium of some funiculi is broken. They are commonly seen by surgeons dealing with trauma (17). The size of the neuroma depends on the number of funiculi severed and the distance between the severed funiculi. Sometimes, they present a challenging surgical problem because a portion of the nerve’s functional status is preserved, and intervention risks downgrading that while attempting repair of the injured segments.
Neuromas after Nerve Repair
Sunderland has shown that neuroma after nerve repair is more common in severed nerves that contain small, widely separated funiculi and/or in nerves where the injury has occurred at a level where the cut ends have dissimilar funicular patterns, as when a segment of nerve has been excised or destroyed (11,17). But there is probably some neuroma formation at each repair site.
Neuromas in Completely Severed Nerves
Neuromas in completely severed nerves are the basis for classically described neuromas. They form at the proximal ends of all discontinuous peripheral nerves.
Amputation Stump Neuromas
Amputation stump neuromas are basically the same as those seen in the completely severed nerve, with three important features: The proximal segment is in close association with the surface tissue of the amputated part and therefore is subject to increased scar formation and to more frequent and repeated trauma. The amputation stump neuroma may become so painful that the amputation stump or the whole part becomes useless. The amputation stump neuroma requires meticulous and timely surgical management to prevent functional loss of the injured extremity secondary to disabling pain (19).
Pathogenesis of Neuroma Formation
The effect of any trauma (laceration, crush, stretch, contusion, radiation) applied to a nerve varies with the magnitude as well as with the duration of the trauma (13).
Depending on the size of the fiber and the topographic location of the fiber within the nerve trunk, the nerve fibers show varying susceptibility to compression. Large fibers are also more vulnerable to compression and ischemia than are small fibers (13,22).
Fibers situated superficially in a fascicle are more vulnerable to compression than are fibers located more centrally in the fascicle (23).
Similarly, nerve fibers within superficial fascicles are affected more than fibers in more deeply located fascicles. Also, the amount of connective tissue in a nerve trunk may influence the effects of compression (e.g., several small fascicles embedded in a larger amount of epineurium are less vulnerable to compression than are large fascicles in a small
P.850

amount of epineurium). This is shown by the fact that the amount of connective tissue in all parts of the nerve is more developed where the nerves are subjected to trauma (e.g., around joints and in superficial nerves) (11).
Although the effect of compression as a chronic example of nerve injury is well studied, data dealing with radiation are limited. Nevertheless, peripheral nerve appears to be a dose-limiting normal tissue in the clinical application of irradiation. Clinically significant peripheral nerve injury will most likely occur if the dose of irradiation exceeds 20 Gy (24). Tolerance of peripheral nerve also depends on the volume of irradiated nerve and surrounding soft tissues. Evaluation of the histomorphometric findings shows a significant decrease in nerve fiber density, particularly in the central portion of the nerve. Electron microscopic analysis has shown an increase in microtubule density and neurofilament accumulation in axons of irradiated nerves, whereas changes in myelin were not present (24).
Findings are suggestive of radiation-induced hypoxia resulting in axon damage and subsequent nerve fiber loss as a mechanism of late radiation injury to the peripheral nerve (25).
The pathogenesis of nerve injury for crushes, cuts, and tears is much different than compression or radiation. It is clear that transection of a peripheral nerve not only divides the axon, the nerve of the cell, but it also divides blood vessels and therefore results in endothelial cell injury, as well as divides connective tissue, with resultant injury to fibroblastlike cells. The injury to the blood vessel allows neutrophils, myocytes, and lymphocytes, as well as platelets and red blood cells, to enter the wounded nerve and therefore interact in the wound healing of the nerve tissue. Thus, an “inflammatory phase” of wound healing in nerve exists (26).
Immediately after injury to the peripheral nerve, an injury message is transmitted to the cell nucleus. The mechanism of this transmission has been partly elucidated and has been brought to attention primarily by de Medinacelli et al. (27).
Regardless of the mechanism of the message, the axon distal to the site of division undergoes the characteristic chances described by Waller. The interaction between the axon and Schwann cell is altered sufficiently that the Schwann cell undergoes changes from a nutritive and supporting cell to a phagocytic cell. If the axon has been myelinated, the Schwann cell that manufactured the myelin now phagocytizes the myelin, which over a period of 2 or 3 weeks undergoes classic degradation changes.
In the early stages of nerve severance, there is swelling followed by fragmentation of the neurofibrils and separation of the myelin sheaths in the distal segment. All traces of the axons disappear by the second or third week. The myelin fragments, axonal remnants, and debris are removed in the process of phagocytosis by either the active Schwann cells or the tissue macrophages. These actions are partially dependent on macrophage production of transforming growth factor-β1 (28).
This entire process is called wallerian degeneration. The distal axon in the process completely disappears. The remnant of the axon–Schwann cell relationship is only the remaining Schwann cell nucleus and cytoplasm and the basal lamina. This degenerated system has been termed band of Büngner. It is clear now that a network of basal lamina and Schwann cell exists in the degenerated peripheral nerve tissue. Electrical stimulation of the distal segment gives a positive response for 3 to 5 days, depending on the speed of axonal degeneration. Although there is a cellular response in the area of injury with some Schwann cell existence and fibroblast proliferation, the response is much less than that in the proximal stump, and the nodule that forms is always smaller than neuroma. This nodule is called the glioma (11,17,19).
Proximal to the nerve transection, the mechanism of axoplasmic transport continues unimpeded. Within 24 hours of the injury, the proximal nerve has begun regenerative changes. At the proximal node to the transection—i.e., for a mechanism of injury that does not cause a significant degree of proximal degeneration such as a transection or crush versus an avulsion—the axon develops multiple axoplasmic sprouts. In this internode region, there is no myelin, and thus the sprouts exit here. The terminal end of each of these sprouts is a growth cone. The growth cone actively seeks an appropriate distal site through movement (18,26,29).
The proximal single axon may yield as many as 10 to 15 sprouts (30). This regenerating unit is guided distally by a combination of the forces of contact guidance and neuro-tropism, and the axon sprout in its growth cone has an affinity for laminin, type 4 collagen, and other subunits (12,28,31,32,33 and 34).
If contact guidance and neurotropism do not work properly, the majority of the regenerating axons that have not reached an appropriate distal target degenerate. The Schwann cell of each involved endoneurial tube in the proximal stump of the severed nerve proliferates, as do the fibroblasts of the endoneurium, perineurium, and epineurium. This process continues unchecked. As a result, axons zigzag through the tissue in a totally disorganized fashion. They branch irregularly and form whorls, spirals, and convolutions. With time, progressive fibrosis converts an originally soft nodule into a firm, hard one (18,20,32). Scar is deposited in concert with this progressive fibrosis and results in the formation of a mass or nodule that physically represents the neuroma (12).
Several parameters are believed to affect the size of the neuroma. The size of the neuroma primarily depends on the extent of axonal ingrowths but is also influenced by the number of fibroblasts, Schwann cells, and blood vessels present. Neuromas tend to be larger closer to the cell body because axoplasmic flow is more intense in those regions. Larger neuromas are seen in nerves with more connective tissue and widely spaced funiculi and in situations of associated soft tissue injury (11). The degree of nerve fiber randomness
P.851

is usually proportional to neuroma size; smaller neuromas tend to have more regularly directed axonal direction (35). Axonal growth is more active the closer the lesion is to the neuron. Therefore, neuromas tend to be larger in nerves injured proximally and smaller when injured peripherally.
Size and consistency of the neuroma are influenced by steroid injection; movement; presence of foreign bodies; and repeated irritation such as pressure, friction, or repeated trauma (17,19,36,37 and 38).
Blood supply does not appear to influence neuroma growth, but general nutritional status seems to affect the size of the neuroma (11,35). The size of the neuroma does not correlate with the symptoms produced by the neuroma.
Recently, investigators have made some new findings dealing with neuroma and its formation. Badalamente et al. showed the presence of myofibroblasts in neuromas. Myofibroblasts increase in the scar tissue from 2 to 6 months after injury and then decrease as the collagen content of the scar increases (14). Molecular biologic techniques show increased RNA production within neuronal cell body, which seems to be correlated with the increased axoplasmic production and transport (39). Glycosaminoglycan content of neuromas is found to be higher than in uninjured nerve connective tissue (14). Procollagen type I mRNA is significantly elevated in the epineurium after crush injury and in endoneurial tissues after transection (40,41). Unique mRNA species are also found in distal nerve segments after injury (42).
Incidence
The incidence of chronic nerve injuries varies on the basis of what kind of chronic injury it is (compression, radiation, direct nerve injury, or unknown etiology). Compression may affect various intraneural tissue components in different ways. Based on the kind of injury, the true incidence for compression injuries is difficult to extract. As the effect of irradiation on nerves is time and dose-limited, it is also difficult to define a true incidence of radiation injury on nerves in the upper extremity. Therefore, we mainly discuss neuroma and its incidence and evaluation.
Neuroma formation was first described by Odier in 1811 (17). A true incidence for neuroma is difficult to define. Many neuromas are reported as relatively asymptomatic, but 20% to 30% are accepted to be painful regardless of what technique of local treatment has been used (43). However, Fisher and Boswick reported a very low incidence of neuroma formation. In their study, four patients demonstrated painful amputation stumps in 144 digital amputation cases. They concluded that this low incidence of painful neuromas and amputation stumps was due to the positive postoperative effort to send patients back to work as soon as possible, allowing them to perform their own therapy and thereby minimizing their disability and tendency to develop pain problems (44).
EVALUATION
The approach to neuromas in the hand requires familiarity with the basic aspects of nerve regeneration as well as the neurophysiology of pain mechanisms.
Pain is usually the presenting symptom and the neuroma may be exquisitely sensitive to touch or to percussion. How an individual perceives “pain” depends on a complex interplay of physiologic occurrences and psychological adaptations.
After an injury or surgery, abnormally intense and persistent pain that is not associated with actual or impending tissue damage may prolong or prevent recovery and impact negatively on health-related quality of life (45). Perception of pain is complex and is dependent on the initiating event, afferent input, efferent modulation, and cortical interpretation. Painful events secondary to cellular damage produce a secondary inflammatory cascade. This includes the activation of pain receptors that input through the dorsal horn of the spinal cord to higher cortical centers (46).
Clinical Evaluation
When evaluating patients with localized pain, a careful review of the history and psychological stability should be combined with a careful and thorough evaluation of the peripheral nerve status. In some patients, especially in workers’ compensation cases, it is believed to be difficult to determine if the neuroma is indeed the major cause of the patient’s pain. In such cases, the patient’s work status and previous work record should be included in the evaluation (17,47).
History
The most common site of painful neuromas is an amputation stump, but stump-type neuromas also occur when sensory cutaneous nerves are severed. Therefore, history should also include previous or current peripheral nerve injury.
Injuries to peripheral nerves are common in all forms of upper extremity trauma, including lacerations, fractures, dislocations, ligamentous tears, and crush or amputation injuries. Common causes include falls and collisions, including athletic competition, motor vehicle accidents, penetrating trauma, and industrial accidents.
When considering the history of someone with neuroma pain, it is important to clarify the type of pain the patient describes. Is the injured area painless or painful, is the pain annoying, does it interfere with some or most of the patient’s activities, or is the patient completely incapacitated with pain (47)? The pain from a neuroma is described as burning, sharp, or aching in nature. A trigger area can usually be identified that is the central focus for the pain stimulus. Some patients with neuromas may even develop reflex sympathetic dystrophic changes, and other patients
P.852

with partial nerve injuries may exhibit a full-blown causalgialike syndrome (18).
Some patients seem to have severe debilitating pain symptoms, whereas others with the same injury remain completely asymptomatic. Therefore, patient specificity appears to be a dominant factor in the pain mechanism. It is believed that a patient’s personality and neuroma symptoms may have a mutual relationship. Patients who tend to be labile with passive-aggressive personalities seem to develop symptomatology, or perhaps their personality changes are secondary to the injury and its sequelae (18).
Physical Examination
When evaluating patients with localized pain, a complete physical examination helps to identify the nature of pain as well as the primary lesion. All severed nerves eventually form neuromas at their stumps, but only a small percentage of neuromas become painful. It is estimated that roughly 5% of all patients admitted to level 1 trauma centers have peripheral nerve injuries (48).
Development of pain in neuroma is related to the location of the nerve that is injured. Digital nerves usually have painful neuromas at the site of injury and/or repair. A painful neuroma developed in a field of previous surgery should suggest the possibility of iatrogenic nerve injury. In that sense, exploration of a neuroma within the dermal scar of an extremity stump or a dense scar associated with the zone of injury can be helpful. The proximity of the painful area to an ischemic environment as well as other location features should be evaluated.
Sometimes, a palpable mass that is tender to touch is present. Pressure on this mass recreates the symptoms of neuroma. The pain produced is described as burning, sharp, or aching in nature. Direct tapping over the nerve can usually identify a trigger area referred distally (Tinel’s sign).
Electrodiagnostic Evaluation
Electrodiagnostic evaluation of chronic nerve injuries requires an understanding of the classification, pathophysiology, and electrodiagnosis of the lesions. This evaluation is also critical to the appropriate diagnosis, localization, and management of nerve trauma in the hand.
The complexity of chronic nerve injury implies the necessity of further diagnostic evaluation. In this section, we discuss the generalized chronic nerve injuries, which include neuroma irradiation and compression-related chronic nerve injuries.
The first step in electrodiagnostic evaluation includes peripheral nerve conduction studies (NCSs) and needle electromyography (EMG). Before the study, skin temperature, presence of callused or wrinkled skin, edema, excess perspiration, and patient’s cooperation should be checked. These factors can all make the study technically more difficult (49).
A peripheral NCS consists of percutaneous delivery of a depolarizing current over a peripheral motor, sensory, or mixed nerve with recording of a compound motor or sensory nerve action potential at a measured distance away from the stimulus through recording electrodes. The electrodes are placed over a muscle belly or nerve. A depolarized nerve carries current in both directions, which is called an orthodromical and antidromical current. This elicited response can be picked up either distally or proximally. The NCS gives two different types of data. The first is motor and sensory conduction velocity, which allows assessment of integrity of the myelin sheath. The second is amplitude of the responses, which can indicate the number of functionally conducting axons in the nerve (49). A complete or partial conduction block can be detected by an NCS. But it should not be forgotten that a wallerian degeneration may take 5 to 7 days to fully develop, and until that point the nerve may remain electrically excitable distal to the injury (50).
In EMG, the basic component is the motor unit action potential, which is a triphasic spike representing the electrical activity of all the muscle fibers from a given motor neuron. EMG is recorded through a recording needle inserted directly in the muscle, and it is an extracellular recording from many muscle fibers. On EMG study, the patient volitionally contracts the muscle being studied, which triggers action potentials. Changes in the shape or firing patterns of action potentials allow assessment of the integrity of the nerve supplying that muscle.
Electrodiagnostic findings help to classify nerve injuries, and such classification has prognostic significance. One of the aforementioned classification systems (Seddon) of peripheral nerve traumatic injuries includes neuropraxia, axonotmesis, and neurotmesis.
The electrodiagnostic finding in the case of neuropraxia is simply referred to as “conduction block.” The nerve distally conducts normally. The cause of conduction block is thought to be a focal demyelination or ischemia. The compound muscle action potential changes immediately after injury. A smaller waveform is noted proximal to the site of blockage. The sensory nerve action potential shows changes that are similar to the compound nerve action potential after neuropraxia. A focal conduction block at the site of the lesion with preserved distal amplitude implies a neuropraxia.
Changes on needle EMG are in recruitment and occur immediately after injury. Although there is no motor unit action potential after a complete lesion, there are reduced numbers of motor unit action potentials firing more rapidly than normal in incomplete neuropraxic lesions.
Axonotmesis describes the situation in a nerve in which the axons and their myelin sheaths are disrupted but the surrounding stroma remains partially or fully intact. In neurotmesis, a nerve has either been completely severed or is so markedly disorganized by scar tissue that axonal regrowth is impossible. The diagnosis of complete axonotmesis and that of complete neurotmesis appear the same based on an electrophysiologic
P.853

study because the difference between these types of lesions lies in the integrity of the supporting structures, which have no electrophysiologic function. A few days after axonotmesis, the compound nerve action potential and motor conduction studies are the same as in a neuropraxic lesion. The consecutive compound nerve action potential study shows a progressive decline in amplitude over the first 5 days, and wallerian degeneration occurs in all motor fibers by approximately 5 to 7 days postinjury (50). Immediately after axonotmesis, sensory nerve action potential appears the same as that seen in a neuropraxic lesion. Nerve segments distal to the lesion remain excitable and demonstrate normal conduction, whereas proximal stimulation results in an absent or small response. Neuropraxia and axonotmesis therefore cannot be distinguished until sufficient time has passed for wallerian degeneration to occur in all sensory fibers. This takes approximately 11 days postinjury, which is longer than in the motor fibers.
Needle EMG demonstrates fibrillation and positive sharp waves a few days after an axon-loss lesion. These waves occur approximately 2 to 3 days before fibrillations (51). Onset of fibrillation potentials after the original nerve injury depends on the length of the distal nerve stump. For very distal injuries, it may take 10 to 14 days for fibrillations to develop. In complete lesions, the earliest needle EMG finding for axonal regrowth is the presence of small, long-duration, polyphasic, often unstable motor unit potentials (52). These potentials reflect the beginning of axonal regrowth as well as new neuromuscular junctions. This is the earliest evidence of reinnervation, which usually precedes the onset of clinically evident voluntary movement (52). In lesions such as partial axon loss lesions as in axonotmesis, recovery depends on axonal sprouting and regeneration. Briefly, there is some early recovery followed possibly by a later recovery if or when regeneration axons reach their end-organs. In this scenario, the amplitude of the compound nerve action potential may provide a guide to prognosis. In complete axonotmesis and neurotmesis, prognosis is the worst. It is impossible to know the degree of nerve injury by electrodiagnostic evaluation. The only way to define the status of nerve injury in complete axon loss is by surgical exploration or by the evidence of early reinnervation after the lesion. Magnetic resonance neurography, a newer noninvasive technique, is also suggested for visualizing the integrity of nerves (49).
Lesions that have a mixture of axon loss and conduction block are more challenging. These injuries have two or more phases of recovery that include both nerve regeneration and hypertrophy of the existing innervated muscle fibers. Therefore, the findings of electrophysiology show the characteristics of the period that is involved.
Localization and assessment of severity of nerve injury in the upper extremity may be complicated. Two methods are used for localization: (a) focal slowing or conduction block on NCSs and (b) pattern of denervation on needle EMG (49). It may be difficult to localize a pure axonotmesis or neurotmetic lesion using nerve conductive study due to loss of fastest fibers in nerve. The second method to determine the site of nerve injury is needle EMG. In fact, there are also a number of potential problems with this approach. The inconsistency of branching and innervation for muscles, the presence of direct muscle trauma, and partial nerve lesions are the most worthy of consideration (51,52).
Electrodiagnostic evaluations also allow objective monitoring of the course of recovery. The studies at 3 to 4 weeks after injury provide a much better diagnostic as well as prognostic database. The findings dealing with recovery may include resolution of conduction block, changes in motor unit firing patterns (e.g., recruitment frequency and morphology), and resolution of abnormal rest activity (49).
Electrodiagnostic techniques may also be applied in the operating room during peripheral nerve surgery. Somatosensory evoked potentials, EMG as spontaneous or free-running electromyographic recordings or as triggered EMG, and NCSs may all be used during all phases of surgery (53,54,55 and 56). Intraoperative neurophysiologic monitoring provides the ability to monitor nerve function, guide dissection, identify neural elements, and assess nerve function. Intraoperative studies confirm preoperative observations and help the surgeon evaluate the functional status of the nerves in question. Therefore, it is stressed that the success of intraoperative electrophysiologic studies is highly dependent on the interaction between the neurophysiologist and the surgeon.
As the visual inspection does not always discern the correct anatomy, neurophysiologic monitoring may protect nerves from risk attributable to manipulation during exposure and dissection of chronic nerve injuries. The goals of the application of electrodiagnostic techniques in surgery can be summarized as the following:
  • To monitor the approach and manipulation of nerves
  • To guide dissection, particularly in scarred tissue in which visual inspection does not always reveal the location of functioning fascicles
  • To identify and localize specific nerves, which in injured and scarred tissue may be difficult to determine
  • To assess nerve function to determine which neural elements are functional
These goals enhance the surgeon’s ability to accurately identify nerves, avoid injury, and effectively treat the defect.
Magnetic Resonance Nerve Imaging
Complete history physical examination and electrophysiologic evaluation help to diagnose most chronic nerve injuries. Nevertheless, some lesions are difficult to classify. Lack of specificity of anatomic definition is thought to be the most limiting factor to diagnose the rest. Ultrasonography and computed tomography scans can give more information about the images of peripheral nerve, but their roles clinically are minor (57,58). With recent technological
P.854

advances in magnetic resonance imaging (MRI), it is possible to make direct magnetic resonance visualization of most peripheral nerves in the upper extremity. The use of specially designed phased-array surface coils to obtain high-resolution images makes magnetic resonance able to visualize nerves in their anatomic position (59). Therefore, MRI can play an adjunctive role in equivocal cases in which symptoms are not classic or NCSs are ambiguous. Identifying focal compression of the peripheral nerves in the upper extremity with MRI could remove ambiguity about the etiology of chronic nerve injury. MRI can determine whether a mass is intrinsic to a nerve or extrinsic to it (60). MRI can also be helpful to identify posttraumatic chronic nerve injuries by demonstrating posttraumatic lesions that might interfere with the healing of axonotmetic injuries, such as neuromas.
The major advance in MRI is believed to be the transition from anatomic to physiologic imaging in the future. These might prove valuable in peripheral nerve injuries. Distinguishing an axonotmetic injury from a neurotmetic injury might be possible, which is sometimes difficult to do with electrophysiologic studies (59).
TREATMENT
Prevention
Neuromas can be painful and physically disabling. Among the many methods of treatment available, the best treatment of painful neuromas is prevention. Therefore, prophylaxis against neuroma formation is of vital importance in the surgical management of wounds. For stump neuromas, this is most conveniently accomplished when each nerve end is identified, dissected from the surrounding soft tissue, and allowed to retract into proximal healthy tissue after amputation (47,61). A large number of painful neuromas can be prevented in this way. Similarly, great care should be taken with all surgical procedures in the upper extremity during which, in the process of exploration, local nerves might be damaged. After cutting a nerve for whatever reason, a surgeon should be sure not to locate a neuroma near or within the suture line or area of soft tissue injury. Neuromas or the cut nerve ends that carry the possibility to form a neuroma should be kept away from prehensile areas and from areas where mechanical stimulation will be frequent. Careful postoperative management is required to facilitate rehabilitation, a return to work, and control of pain symptoms. Education of the patient immediately after surgery of the nature of the Tinel’s sign and sensitivity associated with a neuroma can be helpful. Patients should be taught about desensitization activities so that patients understand and accept some of the discomfort that may be associated with the neuroma. When a progressive pain syndrome is developing, early intervention and physical rehabilitation are essential (18,47).
Nonoperative Treatment
Before any surgical intervention, conservative therapies such as desensitization and physical therapy should be applied. As there is no procedure that is completely and consistently successful in preventing neuroma formation, the literature is extensive on this subject (11).
Physical therapy is accepted to be an important adjunct in the management of patients with chronic nerve injuries. Active and passive range of motion, adaptive modalities including transcutaneous nerve stimulators, contrast baths, hydrotherapy, and continuous passive motion may be helpful (62,63 and 64).
Physiotherapy techniques such as tapping, percussion, massage, and active physiotherapy are also believed to help in alleviation of neuroma symptoms. But no controlled studies have proved their effectiveness.
Transcutaneous vibratory stimulation has progressed in technical sophistication and effective application. It is used for chronic pain problems. It has been shown to reduce pain and to improve dexterity and strength when used twice a week for 1 month (65). It is used as a last-resort modality but may also be useful for certain acute pain problems. Different ranges of pulse rate with pulse duration and amplitude of current are offered. It is stated that sensorial impulses rather than motor stimulation provide desensitization. It is also stated that the therapeutic benefit of the method could be clearly differentiated from the natural healing course because no significant improvement had occurred spontaneously over many years.
Another way of controlling neuroma pain is sympathetic blockade with guanethidine. Guanethidine replaces norepinephrine from the endings of sympathetic nerves. Therefore, it produces a continuous sympathetic blockade lasting from several days to several weeks. It possibly binds to the norepinephrine storage vesicles and releases the substance from the nerve endings. In this technique, an intravenous needle is inserted into a superficial peripheral nerve of the limbs to be blocked. The limb is exsanguinated by means of Esmarch’s bandage. A dose of 15 to 20 mg of guanethidine is injected after being diluted with 20 to 40 mL of normal solution or 0.5% mepivacaine. After the injection, the tourniquet is kept inflated for 10 to 15 minutes to allow the active agent to become fixed within the tissues of the extremity. To check the effectiveness of sympathetic blockade, skin temperature is compared with the temperature of the unblocked extremity (18). Studies also showed that neurolytic blockade with phenol-glycerol might offer an alternative when surgical treatment has not been successful (66).
Surgical Management
The goal of surgical management of chronic nerve injuries and particularly of a painful neuroma is to relieve pain and to restore nerve function. Although many different methods have been described, no universally successful method of surgical
P.855

management has been found. It has been shown that only destruction of the nerve cell body completely inhibits regeneration and therefore neuroma formation (67).
Some authors have proposed a technique for elective digital amputation cases in which an island from the pulp of the amputated finger is isolated while preserving nerve continuity of both collateral pedicles. Once the amputation is completed, in cases of distal amputation, the pulp flap is used to cover the proximal stump, and in cases of ray amputation, the nerve loop is buried after its deepithelization between the adjacent metacarpals. Neuroma formation with this technique is stated as 1% (68). There is currently no available model of comparison for developing better methods of nerve cutting that produces less painful neuromas than those currently in clinical use. Zeltser et al. proposed a model in rats that can be used as a step for development of better neurectomy methods (69). In another study, methods for the treatment of chronic nerve injuries were compared. Four surgical options were described in the study: (a) distal sensory neuromas treated by excision of the neuroma and reimplantation of the proximal nerve into muscle or bone marrow; (b) suspected distal sensory neuromas in which the involved nerve was sectioned proximal to the injury site and reimplanted; (c) proximal neuromas-in-continuity of major sensorimotor nerves treated by external neurolysis; and (d) proximal major sensorimotor nerve injuries at points of anatomic entrapment treated by external neurolysis and transposition. Success rate was the greatest in the last mode of treatment (57%); the least was in the third option (0%) (70).
The extensive list of techniques reflects the fact that no one method reliably prevents formation of a painful neuroma. More than 150 techniques for control of neuroma pain have been described (19).
Herndon stated that the best way to minimize neuroma formation is by a careful repair or grafting to allow the regenerating axons to extend into the distal stump (20).
Algorithm for Surgical Management
Surgical management of the painful neuroma has three basic principles (19):
  • For cases in which an appropriate distal nerve stump and sensory receptors are available, a nerve graft can be used. The nerve graft helps to direct the regenerating nerve fibers from the proximal stump into distal nerve stump and sensory receptors. It is believed to reverse the changes that occur along the peripheral nerve after the division of the nerve as well as to restore function.
  • For cases in which a distal nerve does not exist and restoration of function in the injured nerve is mandatory, vascularized nerve transfers or sensory free tissue transfers can be used to direct the regenerating axons from the proximal stump into end-organs.
  • For cases in which restoration of function in the injured nerve is not mandatory, many different techniques can be used. The methods described under this principle can also be applied when previous treatment failed, when the surrounding soft tissue is not appropriate for a nerve graft, or when the patient refuses treatment based on the first two principles.
There are different surgical options for treatment of neuroma, and their success rates differ from one study to another. Surgical options, indications, results, outcomes, and a review of the literature are discussed in the following sections.
Scar Release and Neurolysis
Releasing the nerve with its neuroma from the surrounding the scar tissue allows the distal neuroma to retract proximally into the healthy soft tissue. Adding internal neurolysis is also recommended for pain relief. This local approach is reported to provide relief in approximately 60% of symptomatic neuromas (18).
The operative technique for this approach is simple and applicable mostly for painful stump neuromas. After neuroma exposure, the surrounding adherent scar tissue is excised by sharp dissection. Further dissection allows the neuroma to retract proximally. The proximal limits of dissection should involve a field of healthy soft tissue that is free of scar and where the new nerve end will be well covered and protected.
Resection
Resection is the choice of treatment of choice for most painful amputation stump neuromas (68,69,70 and 71). It can also be used for neuroma formation acutely if end-to-end repair or grafting is not feasible. If a distal segment exists, resection of both ends to uninjured nerve and grafting are usually the best treatment (see also “Nerve Grafting”). Laborde et al. (72) reported the results of simple resection of 101 painful neuromas of the hand. Reoperative rate was 65%, and ray amputation for symptomatic neuromas resulted in the highest reoperation rate but resulted in little or no subjective improvement. They concluded that there appeared to be a correlation between delayed healing after the initial injury and the severity of neuroma symptoms. They also stated that dorsal translocation of the neuroma consistently resulted in decreased sensitivity without long-term recurrence (72). In another large series of neuromas, 65% of satisfaction or 36.5% excellent results were reported from a single resection. A second simple resection was stated to provide 70% satisfactory results (73).
The operative technique for neuroma resection is similar to the scar-release technique described previously. Under magnification, the nerve involving neuroma is isolated. When dissecting to free the nerve from surrounding soft tissue, care should be taken not to cut any side branch of the nerve or traumatize the nerve trunk. The dissection is limited to a proximal area of healthy soft tissue. This soft
P.856

tissue should be bulky enough to cover and protect the nerve end from future mechanical trauma. Under gentle traction, the nerve end is sectioned in the healthy proximal tissue as proximally as possible (Fig. 3) (74).
FIGURE 3. Simple resection of a painful neuroma is the most common method of management.
Ligation
Ligation is also a technique both for acute prophylaxis and treatment of neuromas. Prevention of painful neuromas with ligation produces unpredictable results. The rationale for ligation of the proximal nerve segments is to obstruct the neural tube by the suture to prevent axonal regeneration beyond the limits of ligature. Some of the studies showed that terminal neuromas formed above the ligation site, and a neuroma was also reported at the nerve end after ligature. The neuroma was found to be smaller but still contained regenerating axons. The fibers below the ligation were histologically nonmyelinated, and it was postulated that the nerve fiber had become smaller to penetrate the site of the ligature (75). Cravioto and Battista performed a clinical study as well as light and electron microscopy studies of painful neuromas (76). It was found that (a) painful neuromas contain large numbers of small-diameter, unmyelinated fibers and (b) a consistent, unstrained growth of perineural cells parallels the constant regeneration of axis cylinders. This resulted in unabated formation of large numbers of “nerve minifascicles” growing in a chaotic fashion. Controversially, it is postulated that the unrestricted growth of perineural cells is an attempt to contain the regeneration of axis cylinders and that the maintenance of perineural integrity by fascicle ligation is important for the relief of painful human neuromas (15,76,77). However, most authors have found ligation unreliable, experimentally as well as clinically (20,73,78,79).
Some authors, after emphasizing the inconsistency of symptoms with neuromas, suggested surgical excision of neuroma along with a combination of funiculectomy, epineural sleeve suture ligation, and silicone capping as the best chance for eradication (80).
FIGURE 4. Diagram of ligature technique. Arrow indicates level of axon cut back. (Reprinted from Yuksel F, Kislaoglu E, Durak N, et al. Prevention of painful neuromas by epineural ligatures, flaps and grafts. Br J Plast Surg 1997;50:183, with permission.)
A combination of ligation along with sympathetic blockade, neurolysis, crushing, capping the nerve (with materials ranging from silicone to metal foil), electrical coagulation, and chemical agents (alcohol, phenol, formaldehyde, gentian violet, tannic acid, collodion) was also reported with mixed success (15,20,78,81,82).
Operative technique is not complex, but ligation of proximal fascicles must be performed carefully. The nerve end to be ligated is dissected free proximally. A nonabsorbable suture is preferred. The ligature is placed around the nerve somewhere between 5 and 10 mm above the cut end and tied tightly enough to seal the funiculi as well as to stop the bleeding. However, circumferential tying should not be so tight as to cut through the nerve. An ample amount of soft tissue should also cover the nerve end to protect from mechanical irritation (Fig. 4).
Epineurial Barrier
Epineural barrier is used for the treatment of painful and/or recurrent painful neuromas. Undamaged epineurium is an impenetrable barrier for regenerating axons, which is believed to prevent neuroma formation (83). Therefore, different techniques have used epineurium as a barrier. Six different types of epineural closures have been described. Those techniques vary from simple epineural closure to complicated closure techniques. Chapple and Corner published their experience with epineural closure in the same year (84,85). Chapple simply advanced the epineurium
P.857

over the end of the severed nerve and sutured it to the opposite epineural edge to close the endoneural tube (84). Although Sunderland demonstrated that normal epineurium prevented the axonal outgrowth, Chapple’s simple closure technique would not have been able to prevent neuroma formation. In the same year, Corner published his experience with the same technique. He also added a wedge excision of distal neural tissue to support the epineural closure, but it also failed (85). Some studies combined epineural closure with crushing and transposition of the stump into muscle; others suggested combining it with funiculectomy. Although the combination technique has better results than former techniques, the overall success rate is still low (73,86). Securing the endoneural tube with a synthetic tissue adhesive is another epineural closure technique. Good results (95%) were reported with this technique (87).
FIGURE 5. Comparison of epineural flap (A) and epineural graft techniques (B). (Reprinted from Yuksel F, Kislaoglu E, Durak N, et al. Prevention of painful neuromas by epineural ligatures, flaps and grafts. Br J Plast Surg 1997;50:183, with permission.)
Capping the nerve end by closure of epineural flaps in the midline has been popularized by Muehleman and Rahimi (88). They suggested that more satisfactory results could be obtained by suturing the terminal epineural flaps onto the distal end of the proximal nerve stump than by epineural ligature alone. Durak et al. also used the same method (89). Their experimental as well as clinical experience revealed that neuroma was also formed in the epineural flap technique. Although closure by epineural flap appears to completely seal off the proximal nerve end, findings dealing with neuroma formation do not make this technique feasible. Yuksel et al. stated that neither a ligature nor capping with epineural flaps sutured in the midline could close off the epineurium completely. The small gaps, especially those between the sutures of the flaps, are considered large enough for regenerating fascicles to grow through and form a neuroma. Therefore, they suggested covering the nerve end with an epineural graft with the sutures placed on the lateral walls of the nerve (83).
These procedures are technically demanding. Under magnification, the epineurium covering the proximal stump is retracted by peeling back over the nerve trunk, and a 10-mm segment of fascicles is sectioned proximally in the epineural flap technique. The two halves of epineural flaps are closed tightly in the midline over the proximal stump with 10-0 nylon sutures. The number of sutures preferred for different-sized nerves is as follows: four sutures in digital nerves, six in radial and ulnar nerves, and eight in median nerves.
In the epineural grafting technique, the fascicles are left untouched. An epineural graft is harvested from the residual nerve tissues of the amputated parts or excised nerve ends that will be repaired. The epineural graft is fashioned as a circular sheet and applied onto the proximal nerve stump to seal off the fascicles completely. The graft is sutured onto the lateral wall of the nerve stump, just 1 mm from the distal margin with 10-0 nylon sutures (four sutures in digital, six in radial and ulnar, and eight in median nerve). After epineural closure of the nerves, the amputation stumps are covered by healthy flaps. Care should be taken to ensure that suture lines do not lie on the nerve endings (Fig. 5).
Transposition of Nerve Ends
Moskowicz’s study indicated that the nerve endings should be placed into an opening in the muscle with a catgut suture after a total resection of the neuroma. It is suggested that muscle grasps the nerve and secures the position of the nerve ending (90). In the transposition technique, the distal end of the proximal nerve stump is turned back on itself and placed in an area devoid of scar. After transposition, a neuroma still develops, but it is presumed to undergo less irritation
P.858

in the new soft tissue bed. Clinical series have been reported that have included patients whose neuromas were embedded in muscle, bone, or nerve tissue (90,91,92,93,94,95,96 and 97). Herndon et al. embedded neuromas into the intrinsic hand muscles, and they preferred transposing the intact nerve end without excision of the developed neuroma (74). Meyer et al. suggested that neuromas might act as a source of spontaneous electrical activity with both C- and A-fiber spontaneous firing. Failure to resect the mature neuroma may in itself be a reason for recurrent neuroma pain (91). If the nerve is twisted during the transposition, small terminal neuromas can be observed in this twisted area (73). A recent prospective study compared intermuscular neuroma transposition with resection for the treatment of Morton’s neuroma. It was concluded that it is unnecessary to excise interdigital neuroma to obtain excellent relief of pain, and transposition of the interdigital neuroma into an intermuscular position produces significantly better long-term results than do the standard resection operations (98).
Many studies have stated similar criteria for an ideal nerve transposition (92,93,94,95,96 and 97): (a) The operation should place the transected end of the sensory nerve well away from an area that is subject to repeated trauma, movement, and mechanical stimulation; (b) there should be no tension on the nerve itself; and (c) the nerve should lie in an area so as to prevent regeneration of the nerve into the skin and minimize the formation of scar tissue about the transected nerve end. These criteria necessitate implanting as well as transposing the nerve ending to the soft tissue. Therefore, most of the studies recommend implantation and transposition simultaneously. Petropoulos and Stefanko suggested dissecting out the central stump of the cut nerve into its longitudinal fascicles and implanting each of the funiculi into the tissue separately (75).
Many authors have explored the concept of resecting the neuroma and implanting the proximal stump into healthy tissues. Up to now, muscle, bone, nerve itself, and vein have been used for implantation.
Muscle Implantation
The earliest report dealing with muscle implantation was published in the beginning of the last century (92). The first experimental study of muscle implantation, reported by Teneff, revealed no trace of connective tissue proliferation or formation of amputation neuroma (99).
Dellon et al. presented their preliminary results about muscle implantation in 1984 (Fig. 6) (90). They suggested that transected cutaneous nerves do not form neuromas after long-term implantation into the muscle. In 1986, Dellon and Mackinnon listed the criteria for muscle implantation. They reported an 82% overall success rate. They postulated that factors predictive of a poorer outcome are digital neuroma and to be operated on previously three or more times for pain relief (92). Evans and Dellon (100) also suggested implanting palmar cutaneous branch of the median nerve into the pronator quadratus muscle for the treatment of painful neuroma after carpal tunnel release. Their results were good or excellent. The same technique was also reported for another series of painful neuroma patients in 1998 (101). In this series, the pronator quadratus muscle proved a suitable site for transposition of sensory nerve ends after resection of painful neuromas in the proximal part of the hand and wrist.
FIGURE 6. Painful neuromas can be managed by transposition into the muscle.
Bone Implantation
Boldrey, in 1943, conducted an experimental study on dogs in which acutely lacerated ulnar nerves were buried into the medullary canal (Fig. 7) (102). The neuroma formation was smaller than those buried outside the bone. Bone implantation is suggested to help restrict neuroma size as well as protect the neuroma from direct irritation (17,102,103,104 and 105). Mass et al. proposed bone implantation technique for those who had persistent or resistant pain (103). They reported 90% good results. After subperiosteal dissection, two holes are created in bone large enough to easily admit the neuroma and to thread the suture for anchoring the nerve in bone. They postulated that the reason that none of the patients had a significant residual Tinel’s sign was because the neuroma was inside the bone and there was no tension on the nerve remaining outside the bone. Other studies with varying satisfactory results were also reported (104,105).
FIGURE 7. Another neuroma management technique is to transpose the sectioned nerve into bone.
P.859

Vessel Implantation
Among the many methods of treatment available, one of the most interesting techniques is implantation of a nerve ending into a vein (106,107). These studies evaluated neuroma prevention with implanting a nerve ending into a vein. Availability and accessibility of veins are reasoned for this technique. Based on the theory that recurrent neuroma formation can be prevented if the cut nerve end is implanted into the lumen of a vein, 14 patients were treated by neuroma excision followed by proximal vein implantation (106). Satisfactory results (80%) were reported in another experimental study. Medial branches of the right femoral nerve of 20 rats were transected at mid-thigh level, and proximal segments were implanted into the intact femoral vein. The epineurium was sutured to the tunica adventitia of the vein. The regenerated nerve fibers were found within the adventitia and muscular wall of vein. No extension of nerve tissue into the vessel lumen and no thrombosis of the vein were reported (107).
Nerve Implantation
Some ideas of profiting from an adjacent healthy nerve have always attracted the surgeons. This approach has generated enthusiasm for including the adjacent nerves to limit possible neuroma formation of a free nerve end. Some authors have preferred to involve the same nerve trunk, whereas others have preferred the adjacent nerves (108,109,110,111,112,113,114,115 and 116). Different techniques have been reported with varying success rates.
Same-Nerve Implantation
Implantation into the same nerve trunk, termed neurocampsis, was presented by Petropoulos and Stefanko (35,75). They described a spiral-shaped neuroma in their experimental study that was twice the size of the control neuromas. In this technique, the cut nerve end is released to reach the same nerve trunk. An epineural window is created at the site where the nerve is transposed, and this nerve flap is sutured to the nerve trunk. This technique was reported with high success rates (20,108,109 and 110). One experimental study evaluated neuroma prevention in an end-to-side repair technique. The proximal segment was looped back to the main nerve, and end-to-side epineural repair was performed. Electron microscopic studies were also performed in the study, and it was concluded that end-to-side repair formed a smaller mass of regenerated nerve tissues (111).
Adjacent Nerve Implantation
A number of different methods have been described to transpose the severed nerve end to an adjacent healthy nerve. These implantations were used in an attempt to direct axonal growth and prevent the development of a neuroma.
In 1904, Longley and Anderson showed that regenerating axons would not develop into endoneurial tubes that are already occupied by an axon (112). This concept suggested to juxtapose cut nerve ends to limit neuroma formation. But Sunderland described the multifuniculated structure of human nerves that are separated by epineural connective tissue (11). Regenerating axons from each nerve grow into and branch within the epineural tissues of the other, forming neuroma-in-continuity.
Harrison reported two patients with finger amputation and persistent pain. Both patients remained free from pain after resection of the neuromas and suturing the two digital nerves together in the palm (113).
Centrocentral Nerve Union
The technique of centrocentral nerve union with autologous transplantation is described as the coaptation of two nerve bundles from the same nerve trunk. It was first described by Gorkisch et al. (114). This technique is also suggested for one nerve if it is split into two fascicles of equal size. In addition to a simple end-to-end repair, a nerve segment is again severed proximally. Thus, a 5- to 10-mm length autograft is artificially created. This autograft is suggested to prevent the axons coming from both nerves to meet in the suture line during the regenerating process. It is believed that maintaining the epineurium when creating the second cut is important to ensure nerve viability. A second microneurorrhaphy is then performed, and the denervated segment of nerve is then a receiving conduit for active fibers that previously would have formed a neuroma (Fig. 8) (47,114).
Kon and Bloem used the centrocentral nerve union with an interposed autologous nerve graft for the treatment of 32 symptomatic neuromas in the fingers of 18 patients. A satisfactory improvement was reported in all 18 patients (115). In this technique, two advantages have been suggested. First, the outgrowth of the axons into the graft may lead to increased pressure, which, instead of enhancing the development of intraperineural neuromas, guides the growing ends of the axons toward each other. A second advantage of this increased intraperineural pressure is the reduction of protein production and axoplasm flow in the neuron, which may also play a role in the inhibition of neuroma development (113,114 and 115). So far, most of the explanations for this phenomenon are still hypothetical.
Some reports dealing with centrocentral union technique usage in lower extremity have yielded high-percentage success rates (116,117).
FIGURE 8. The centrocentral nerve union technique is applicable when transposition or simple resection is not feasible.
P.860

Implantation via End-to-Side Nerve Union
1992 and 1994 studies on end-to-side nerve repair demonstrated axonal passage through alternate pathways (118,119). Thereafter, methods using this repair technique have been popularized. As stated, end-to-side union technique was previously applied within the same nerve (111). Recently, in an experimental study, the use of the end-to-side nerve union technique to prevent and treat painful neuromas was investigated (110). In the experimental model, the author sutured the cut ends of the tibial nerve to the adjacent peroneal nerve in an end-to-side fashion in rats. The proximal end of the tibial nerve was found to form a “nonclassic” neuroma and heal into the peroneal nerve with continuity of the epineurium. Preliminary clinical experience using the same technique in the prevention and treatment of painful neuromas of the superficial radial nerve is also presented in the study (110).
Capping
The goal of proximal neuroma control has inspired a variety of surgical procedures. One is capping the nerve end with different alloplastic materials ranging from silicone to metal foil (78,80,81,83). The severed proximal end of the nerve is sheathed in these numerous materials to diminish neuroma formation. Studies with silicone, polyethylene, methylmethacrylate, rubber, plastic cellophane, Lucite, collodion, vessels, decalcified bone, fascia, placental tissue, dried plasma, epineurium, silver, Vitallium, tantalum, glass, and tin as nerve caps suggested a method of reducing neuroma formation and scar production. With these capping methods, a capsule forms within the cap, encasing the axons (17,19,78,81,120,121,122,123,124 and 125).
None of the studies yielded complete and conclusive results. Inconsistency of results in some of the studies is attributed to the difficulty in using capping material and to foreign body reactions. Success rates are low and unpredictable. Ineffectiveness of these materials made them all historical except for silicone.
Silicone Capping
Silicone capping has been reported to be more effective than any other capping techniques. Swanson et al. (81) evaluated a silicone capping technique of severed nerves in rabbits. Neuroma inhibition was achieved when a length-to-diameter ratio of 5:1 (minimum) to 10:1 (maximum) of the cap was observed. They also noted a smooth, filmy, nonreactive capsule around all nerve ends and caps. They reported a series of 38 neuromata in 18 patients with satisfactory results (81). The possible means by which this technique helps to prevent neuroma formation are the following: (a) Silicone use results in the formation of a fine capsule that is smooth, malleable, and relatively vascular, and this provides a smooth fibrous cover for the nerve end that partially prevents the formation of dense scar tissue surrounding neuroma; and (b) the silicone cap itself affords some element of protection for the tissue within.
In this technique, after resecting the neuroma, a silicone cap that is slightly larger than the nerve is chosen. The length-to-diameter ratio should be a minimum of 5:1. The cap is then secured over the nerve end by using 5-0 nonabsorbable Bunnell—s-type suture.
Coagulation
Many coagulation methods to suppress regeneration of severed nerve ends have been used, including physical and chemical methods.
Physical coagulation methods include laser, heat, electrocoagulation, and freezing (20,75,126). These methods generally are believed to be unsuccessful because of the high capacity of the axons to regenerate. Cryogenic denervation of Morton’s neuromas in the foot has been suggested (127,128). No incidence of neuroma formation or neuritis is reported, but research seems to be insufficient for conclusion.
Prevention of neuroma formation via laser has also been used. Experimental observations of Nd:YAG laser as well as carbon dioxide milliwatt laser need to be evaluated with further studies (129,130).
Chemical coagulation methods include sclerosing agents such as alcohol, phenol, hyalase, pepsin, gentian violet, nitrogen mustard, radioactive isotopes, procaine, osmic acid, tannic acid, picric acid, chromic acid, and hydrochloric
P.861

acid (20,67,78,81,83,123). Based on results, these agents are believed to have no real value in neuronal prevention. Besides these sclerosing agents, steroids were also tried to suppress neuroma formation. Steroids are injected deep enough so that the solution reaches the neuroma and the soft tissue around it. Steroid injection has been advised mostly for digital neuromas. Occurrence of side effects such as fat pad atrophy in hand and foot and local cutaneous changes has been reported. Therefore, injections should be deep enough to prevent leaking into the subcutaneous area (131,132,133 and 134).
Flap Transfer
Brown and Flynn first introduced the concept of covering neuromas with thick flaps 1973 (135). Holmberg et al. generalized the use of flaps to relieve posttraumatic pain in the upper extremity (136). Millesi et al. suggested that gliding fascia is the best tissue to cover an injured nerve (137,138). Jones recommended neurolysis of the involved nerve followed by “wrapping” the nerve with either a pedicled or free flap (139). Foucher et al. advocated coverage of painful digital neuromas with local flaps (e.g., advancement, island flap) and distant flaps (e.g., free “custom-made” toe flaps) (140). They reported 87% excellent or good results. Studies suggesting lumbrical muscle flaps for the treatment of digital neuromas and medial triceps flaps for ulnar neuromas also reported satisfactory results (141,142). It has been suggested that several mechanisms are important to relieve the symptoms of chronic pain (139):
  • Reliable blood supply of flaps may promote revascularization of a devascularized segment of the nerve.
  • Wrapping the nerve with a flap may insulate the nerve from the traction forces of adjacent moving tendons.
  • Bulky flap tissue may cushion the nerve from external pressure on the overlying skin.
  • Well-vascularized flaps result in less fibrosis around a nerve tissue.
Nerve Grafting
The best approach to prevention and therapeutic management of a neuroma is precise alignment of the proximal and distal nerve ends such that all proximal fascicular “outflow” is channeled into the distal endotube. This meticulous alignment usually prevents the formation of neuromas. However, patients exhibit marked discomfort at the site of surgical repair that is not amenable to the usual forms of therapy (18,19). These patients present a difficult task for hand surgeons, and surgical neuroma excision followed by nerve grafting may be the only sensible form of treatment.
The first report of any nerve graft was by Philipeaux in 1870. He interposed a gap of a hypoglossal nerve of a dog with an autologous lingual nerve. The first nerve allograft was reported in 1878, and the first report of clinical outcome of autograft was presented in 1885 (143). Mayo-Robinson repeated the first successful nerve graft in humans when he used an allograft to bridge a 2.5-cm gap in a median nerve (144). In the 1960s, microneurosurgery developed, and Millesi et al. introduced interfascicular grafting and emphasized the importance of good microsurgical technique, including instrumentation, sutures, and magnification in obtaining success with nerve grafts (145). Taylor and Ham reported the first successful vascularized nerve graft in 1976 (146).
Types of Grafts
Full-thickness segments of a major nerve trunk, or trunk grafts, were the first structures used in both clinical and experimental nerve grafting. Because of the central necrosis and total graft dissolution, this practice was short-lived (147). The cable graft began to replace the trunk graft after the disappointing results. Strands of cutaneous nerve were sewn together at both ends and trimmed to form a flat surface equal in diameter to that of the injured nerve. The entire cable was then sewn to the proximal and distal stump. When both adjacent nerves are damaged, one of them may be used as a flap to repair the other, as described by St. Clair Strange (148). In interfascicular nerve grafting, the structures being joined are fascicular groups. The crux of the operation is the matching of fascicular groups in the proximal and distal stumps. With gaps of 4 cm to 6 cm, excellent correspondence of proximal and distal patterns may be obtained (149). Grafting of individual fascicles is another option in grafting. As this approach is more time-consuming, it is rarely performed. Finally, the free vascularized nerve grafting was introduced in an attempt to prevent ischemic graft failure (146).
A variety of materials have been used as substitutes for autogenous nerve in the bridging of nerve gaps. Autologous vein conduits are the most readily available substitutes. The use of allograft nerve as a substitute for bridging autogenous nerve gaps with artificial conduits has recently been the subject of intense experimental investigation.
Sources of Autogenous Nerve Grafts
The sural nerve has become the standard source for grafting large defects in the median and/or minor nerves. Cutaneous nerves of the forearm are often sacrificed to bridge digital nerve defects. The lateral antebrachial cutaneous nerve, the anterior branch of the medial antebrachial cutaneous nerve, and the terminal branch of the posterior interosseous nerve are harvested and used as graft (150).
Nerve Grafting for the Treatment of Neuroma-in-Continuity
The diagnosis of neuroma-in-continuity is usually made months after nerve injury. The progression of Tinel’s sign and
P.862

evidence of functional return are evaluated at monthly intervals; electrodiagnostic testing may also be helpful. Electrodiagnostic testing may provide useful information before clinical recovery is evident. The decision to explore a suspected neuroma-in-continuity is based on failure of recovery or if recovery takes so long in some locations that intraoperative testing is preferable to obligatory end-organ degeneration.
When the decision for surgical exploration has been made, consultation with an electromyographer is necessary. The electromyographer performs the required intraoperative conduction studies.
Broad exposure down to the plane of the nerve is mandatory. This enables placing the stimulating and recording electrodes on nerve segments proximal and distal to the lesion. The electrodes must be spaced at least 4 cm apart to avoid false recording (150). Scarred epineurium is microsurgically removed from the area of the lesion. If the internal epineurium is also scarred, internal neurolysis is performed.
The functioning motor fascicles are identified proximal and distal to the injury site, with electrical nerve stimulation eliciting muscle contraction (151). These motor fascicles are preserved. The electrically silent and nonfunctioning sensory fascicles are divided proximal and distal to the neuroma and reconstructed with autogenous nerve grafts. These nerve grafts bypass the functioning motor portion of the neuroma-in-continuity.
REFERENCES
1. Barr ML, Kiernan JA, eds. The human nervous system, 6th ed. Philadelphia: JB Lippincott Co, 1993.
2. Van Beek AL, Jacobs SC, Zook EG. Examining of peripheral nerves with the scanning electromicroscope. Plast Reconstr Surg 1979;65:309.
3. Allieu Y, Bonnel F. Surgical anatomy of the peripheral nerves. In: Tubiana R, ed. The hand, vol. 3. Philadelphia: WB Saunders, 1988:405–422.
4. Smith JW. Factors influencing nerve repair, collateral circulation of peripheral nerves. Arch Surg 1966;93:433–437.
5. Williams HB, Jabaley ME. The importance of internal anatomy of peripheral nerves to nerve repair in the forearm and hand. Hand Clin 1986;2:689–691.
6. Sunderland S. The intraneural topography of the radial, median and ulnar nerves. Brain 1945;68:243.
7. Jabaley ME, Wallace WH, Heckler FR. Internal topography of major nerves of the forearm and hand. A current view. J Hand Surg [Am] 1980;5:1.
8. Lundborg G, Branemark PI. Microvascular structure and function of peripheral nerves. Vital microscopic studies of the tibial nerve in the rabbit. Acta Microcirc 1968;1:66–88.
9. Lundborg G. The intrinsic vascularization of human peripheral nerves: structural and functional aspects. J Hand Surg [Am] 1979;4:34.
10. Zachary LS, Delon AL, Seiler WA. Relationship of intraneural damage in the rat sciatic nerve to mechanism of injury. J Reconstr Microsurg 1989;5:137.
11. Sunderland S. Nerves and nerve injuries, 2nd ed. Edinburgh: Churchill Livingstone, 1978.
12. Mackinnon SE, Delon AL. Surgery of the peripheral nerve. New York: Thieme Medical Publishers, 1988.
13. Dahlin LB, Lundborg G. The neuron and its response to peripheral nerve compression. J Hand Surg [Br] 1990;15:5.
14. Badalamente MA, Hurst LC, Ellstein J, et al. The pathobiology of human neuromas: an electron microscopic and biochemical study. J Hand Surg [Br] 1985;10:49–53.
15. Battista AF, Craviato HM, Budzilovich GN. Painful neuroma: changes produced in peripheral nerve after fascicle ligation. Neurosurgery 1981;9:589–590.
16. Seddon H. Surgical disorders of the peripheral nerves. Edinburgh: Churchill Livingstone, 1972.
17. Sunderland S. Nerves and nerve injuries. Baltimore: Wil-liams & Wilkins, 1968.
18. Williams HB. The painful stump neuroma and its treatment. In: Terzis JK, ed. Microreconstruction of nerve injuries. Philadelphia: WB Saunders, 1987:161–172.
19. Nath RK, Mackinnon SE. Management of neuromas in the hand. Hand Clin 1996;12:747–748.
20. Herndon JH. Neuromas. In: Green DP, Hotchkiss RN, Pederson WC, eds. Green’s operative hand surgery, 4th ed. New York: Churchill Livingstone, 1999:1470–1472.
21. Reference deleted.
22. Dahlin LB, Shyu BC, Danielsen N, et al. Effects of nerve compression or ischemia on conduction properties of myelinated and non-myelinated nerve fibers. An experimental study in the rabbit common peroneal nerve. Acta Physiol Scand 1989;136:97–105.
23. Powell HC, Myers RR. Pathology of experimental nerve compression. Lab Invest 1986;55:91.
24. Vujaskovic Z. Structural and physiological properties of peripheral nerves after intraoperative irradiation. J Peripher Nerv Syst 1997;2:343–349.
25. Hao JX, Blakeman KH, Yu W, et al. Development of a mouse model of neuropathic pain following photochemically induced ischemia in the sciatic nerve. Exp Neurol 2000;163:231–238.
26. Dellon AL. Wound healing in nerve. Clin Plast Surg 1990;17:545–552.
27. de Medinacelli L, Freed WJ, Wyatt RT. An index of the functional condition of rat sciatic nerve based on measurements made from walking tracks. Exp Neurol 1982;77:634.
28. Einheber S, Hannocks MJ, Metz CN, et al. Transforming growth factor-beta 1 regulates axon/Schwann cell interactions. J Cell Biol 1995;129:443–458.
29. Van Hooff COM., Oestreicher AB. The nerve gap: suture under tension versus graft. Plast Reconstr Surg 1975;56:96.
30. Morris J, Hudson A, Weddell G. A study of degeneration and regeneration in the divided rat sciatic nerve based on electron microscopy. I. The traumatic degeneration of myelin in the proximal stump of the divided nerve. Z Zellforsch Mikrosk Anat 1972;124:76.
31. Letourneau PC. Neurite extension by peripheral and central nervous system neurons in response to substratum-bound fibronectin and laminin. Dev Biol 1983;95:212.
32. Dubovy P, Svizenska I, Klusakova I, et al. Laminin molecules in freeze-treated nerve segments are associated with migrating Schwann cells that display the corresponding alpha-6 beta-1 integrin receptor. Glia 2001;33:36–44.
P.863

33. Mligliche N, Kitada M, Ide C. Grafting of detergent-denaturated skeletal muscles provides effective conduits for extension of regenerating axons in the rat sciatic nerve. Arch Histol Cytol 2001;64:29–36.
34. Yuksel F, Karacaoglu E, Guler MM. Nerve regeneration through side-to-side neurorrhaphy sites in a rat model: a new concept in peripheral nerve surgery. Plast Reconstr Surg 1999;104:2092–2098.
35. Petropoulos PC, Stefanko S. Experimental studies of posttraumatic neuromas under various physiologic conditions. J Surg Res 1961;1:235–240.
36. Boyes JH. Bunnel’s surgery of the hand, 3rd ed. Philadelphia: JB Lippincott Co, 1956:426.
37. Campbell JB, Bassett CAL, Girado JM, et al. Application of monomolecular filter tubes in bridging gaps in peripheral nerves and for prevention of neuroma formation. A preliminary report. J Neurosurg 1956;16:635.
38. Davis L, Perret G, Hiller F. Experimental studies in peripheral nerve surgery. Effect of injection on regeneration and functional recovery. Surg Gynecol Obstet 1945;81:302–308.
39. Engh CA, Schofield BH. A review of the controlled response to peripheral nerve injury and its significance in nerve regeneration. J Neurosurg 1972;37:198.
40. Salonen V, Lehto M, Vaheri A, et al. Endoneurial fibrosis following nerve transection. Acta Neuropathol (Berl) 1985;67:315–321.
41. Siironen J, Sandberg M, Vuorinen V, et al. Collagen type I and III and fibronectin expression after transection of peripheral nerve: reinnervation compared to denervation. Lab Invest 1992;67:80–87.
42. De Leon M, Welcher AA, Suter U, et al. Identification of transcriptionally regulated genes after sciatic nerve injury. J Neurosci Res 1991;29:437–448.
43. Nelson AW. The painful neuroma. The regenerating axon versus the epineural sheath. J Surg Res 1977;23:215–221.
44. Fisher GT, Boswick JA Jr. Neuroma formation following digital amputations. J Trauma 1983;23:136–142.
45. Koman AL, Smith T, Smith BP, et al. The painful hand. Hand Clin 1996;12:757–764.
46. Raya SN, Meyer RA, Campbell JN. Peripheral mechanisms of somatic pain. Anesthesiology 1988;68:571.
47. Van Beek AL. Management of nerve compression syndromes and painful neuromas. In: McCarthy JG, ed. Plastic surgery, vol. 7. Philadelphia: WB Saunders, 1990:4846–4858.
48. Noble J, Munro CA, Prasad VSSV, et al. Analysis of upper and lower extremity peripheral nerve injuries in a population of patients with multiple injuries. J Trauma 1998;45:116–122.
49. Carter GT, Robinson LR, Chang VH, et al. Electrodiagnostic evaluation of traumatic nerve injuries. Hand Clin 2000;16:1–12.
50. Claudry V, Cornblath DR. Wallerian degeneration in human nerves. Serial electrophysiological studies. Muscle Nerve 1992;15:687–693.
51. Kraft GH. Fibrillation amplitude and muscle atrophy following peripheral nerve injury. Muscle Nerve 1990;13:814–821.
52. Dorfman LJ. Quantitative clinical electrophysiology in the evaluation of nerve injury and regeneration. Muscle Nerve 1990;13:822–828.
53. Buchberger W, Judmaier W, Birbamer G, et al. Carpal tunnel syndrome: diagnosis with high-resolution sonography. AJR Am J Roentgenol 1992;159:793–798.
54. Jessurun W, Hillen B, Zonneveld F, et al. Anatomical relations in the carpal tunnel: a computerized tomographic study. J Hand Surg [Br] 1987;12:64–67.
55. Jarvik JG, Kliot M, Maravila KR. MR nerve imaging of the wrist and hand. Hand Clin 2000;16:13–24.
56. Ikeda K, Haughton V, Ho K-C, et al. Correlative MR anatomic study of the median nerve. AJR Am J Roentgenol 1996;167:1233.
57. Oberle J, Antoniadis G, Rath S, et al. Value of nerve action potentials in the surgical management of traumatic nerve lesions. Neurosurgery 1997;41:1337–1342.
58. Ochiai N, Nagano A, Sugiako H, et al. Nerve grafting in branchial plexus injuries. Results of free grafts in 90 patients. J Bone Joint Surg Br 1996;78:754–758.
59. Tiel R, Happel L, Kline D. Nerve action potential recording method and equipment. Neurosurgery 1996;39:103–108.
60. Slimp JC. Intraoperative monitoring of nerve repairs. Hand Clin 2000;16:25–36.
61. Wu J, Chiu DT. Painful neuromas: a review of treatment modalities. Ann Plast Surg 1999;43:661–667.
62. Long DM. Electrical stimulations for relief of pain from chronic nerve injury. J Neurosurg 1973;39:718.
63. Melzack R. Prolonged relief of pain by brief, intense transcutaneous somatic stimulation. Pain 1975;1:357.
64. Richlin DM, Corron H, Dawlingson JC, et al. Reflex sympathetic dystrophy: successful treatment by transcutaneous nerve stimulation. J Pediatr 1978;93:84.
65. Spicer C, Kalut G. Rapid relief of a painful, long-standing, posttraumatic digital neuroma treated by transcutaneous vibratory stimulation (TVSS). J Hand Ther 1996;9:47–51.
66. Kirvela O, Nieminen S. Treatment of painful neuromas with neurolytic blockade. Pain 1990;41:161–165.
67. Guttman L, Medawar PB. Chemical inhibition of fiber regeneration and neuroma formation in peripheral nerves. J Neurol Psychiatry 1942;5:130.
68. St. Laurent JY, Duclos L. Prevention of neuroma in elective digital amputations by utilization of neurovascular island flap. Ann Chir Main Memb Super 1996;15:50–54.
69. Zeltser R, Beilin B, Zaslansky R, et al. Comparison of autotomy behavior induced in rats by various clinically-used neurectomy methods. Pain 2000;89:19–24.
70. Burchiel, KJ. Johans TJ, Ochoa J. Painful nerve injuries: bridging the gap between basic neuroscience and neurosurgical treatment. Acta Neurochir Suppl 1993;58:131–135.
71. Eaton RG. Painful neuroma. In: Omar GE, Spinner M, eds. Management of peripheral nerve problems. Philadelphia: WB Saunders, 1980.
72. Laborde KJ, Kalisman M, Tsai M. Results of surgical treatment of painful neuromas of the hand. J Hand Surg [Am] 1982;7:190–193.
73. Tupper JW, Booth DM. Treatment of painful neuromas of sensory nerves in the hand: a comparison of traditional and newer methods. J Hand Surg [Am] 1976;1:144–151.
74. Herndon JH, Eaton RG, Littler JW. Management of painful neuromas in the hand. J Bone Joint Surg Am 1976;58:369–373.
75. Petropoulos PC, Stefanko S. Experimental observations on the prevention of neuroma formation. J Surg Res 1961;1:241–248.
P.864

76. Cravioto H, Battista A. Clinical and ultrastructural study of painful neuroma. Neurosurgery 1981;8:181–190.
77. Battista A, Cravioto H. Neuroma formation and prevention by fascicle ligation in the rat. Neurosurgery 1981;8:191–204.
78. De Carvalho Pinto VA, Junqueira LCU. A comparative study of the methods for prevention of amputation neuroma. Surg Gynecol Obstet 1954;99:492–494.
79. Gonzalez-Darder J, Barbera J, Alamo J, et al. Fascicular ligation in the prevention and treatment of painful terminal neuroma: an experimental study in the rat. Neurosurgery 1987;21:215–217.
80. Williams HB. The painful stump neuroma and its treatment. Clin Plast Surg 1984;1:79–84.
81. Swanson AB, Boeve NR, Lumsden RM. The prevention and treatment of amputation neuromata by silicone capping. J Hand Surg [Am] 1977;2:70.
82. Huber GC, Lewis D. Amputation neuromas. Their development and prevention. Arch Surg 1920;1:85–113.
83. Yuksel F, Kislaoglu E, Durak N, et al. Prevention of painful neuromas by epineural ligatures, flaps and grafts. Br J Plast Surg 1997;50:182–185.
84. Chapple WA. Prevention of nerve bulbs in stumps. BMJ 1918;1:399.
85. Corner EM. The structure, forms and conditions of the ends of divided nerves: with a note on regeneration neuromata. Br J Surg 1918;6:273–278.
86. Munro D, Mallory GK. Elimination of the so-called amputation neuromas of divided peripheral nerves. N Engl J Med 1959;260:358.
87. Martini A, Fromm B. A new operation for the prevention and treatment of amputation neuromas. J Bone Joint Surg Br 1989;71:379–382.
88. Muehleman C, Rahimi F. Effectiveness of epineural barrier in reducing axonal regeneration and neuroma formation in the rat. J Foot Surg 1990;29:260–264.
89. Durak N, Yuksel F, Kislaoglu E. Effectiveness of epineural barriers as a flap and free graft in reducing neuroma formation in the rat following sciatic nerve severance. Eur J Plast Surg 1995;18:272–275.
90. Dellon AL, Mackinnon SS, Pestronk A. Implantation of sensory nerve into muscle: preliminary clinical and experimental observations on neuroma formation. Ann Plast Surg 1984;12:30–40.
91. Meyer RA, Campbell JN, Roya SN. Neural activity originating from a neuroma in the baboon. Soc Neurosci Abstr 1982;8:855.
92. Dellon AL, Mackinnon SE. Treatment of the painful neuroma by neuroma resection and muscle implantation. Plast Reconstr Surg 1986;77:427–436.
93. Mackinnon SE, Dellon AL. Overlap of lateral antebrachial cutaneous nerve and superficial branch of the radial nerve. J Hand Surg [Am] 1985;10:522.
94. Reisman NR, Dellon AL. The abductor digiti minimi muscle flap: a salvage technique for palmar wrist pain. Plast Reconstr Surg 1983;72:859.
95. Dellon AL, Mackinnon SE. Susceptibility of the superficial sensory branch of the radial nerve to form painful neuromas. J Hand Surg [Br] 1984;9:42.
96. Parry CBW. Painful conditions of peripheral nerves. Aust N Z J Surg 1980;50:233.
97. Battista AF. Pain of peripheral nerve origin: fascicle ligation for the prevention of painful neuroma. Adv Pain Res Ther 1979;3:167.
98. Calgrove RC, Huang EY, Barth AH, et al. Interdigital neuroma: intermuscular neuroma transposition compared with resection. Foot Ankle Int 2000;21:206–211.
99. Teneff S. Prevention of amputation neuroma. J Int Coll Surg 1949;12:16.
100. Evans GR, Dellon AL. Implantation of the palmar cutaneous branch of the median nerve into the pronator quadratus for treatment of painful neuroma. J Hand Surg [Am] 1994;19:203–206.
101. Sood MK, Elliot D. Treatment of painful neuromas of the hand and wrist by relocation into the pronator quadratus muscle. J Hand Surg [Br] 1998;23:214–219.
102. Boldrey, E. Amputation neuroma in nerves implanted in bone. Ann Surg 1943;118:1052.
103. Mass DP, Ciano MC, Tortosa R, et al. Treatment of painful hand neuromas by their transfer into bone. Plast Reconstr Surg 1984;74:182–185.
104. Hemmy DC. Intramedullary nerve implantation and other traumatic neuromas. J Neurosurg 1981;54:842–843.
105. Goldstein SA, Sturim HS. Intraosseous nerve transposition for treatment of painful neuromas. J Hand Surg [Am] 1985;10:270–274.
106. Herbert TJ, Filan SL. Vein implantation for treatment of painful cutaneous neuroma. A preliminary report. J Hand Surg [Br] 1998;23:220–224.
107. Low CK, Chew SH, Song IC, et al. Implantation of a nerve ending into a vein. Clin Orthop 2000;379:242–246.
108. Ashley L, Stallings JO. End-to-side nerve flap for treatment of painful neuroma: a 15 year follow up. J Am Osteopath Assoc 1988;88:621–624.
109. Herndon JH, Hess AV. Neuromas. In: Gelberman RG, ed. Operative nerve repair and reconstruction. Philadelphia: JB Lippincott Co, 1991:1525–1540.
110. Al-Qattan MM. Prevention and treatment of painful neuromas of the superficial radial nerve by the end-to-side nerve repair concept: an experimental study and preliminary clinical experience. Microsurgery 2000;20:99–104.
111. Low CK, Chew SH, Song IC, et al. End-to-side anastomosis of transected nerves to prevent neuroma formation. Clin Orthop 1999;369:327–332.
112. Longley JN, Anderson HK. The union of different kinds of nerve fibers. J Physiol 1904;31:365.
113. Wood VE, Mudge MK. Treatment of neuromas about a major amputation stump. J Hand Surg [Am] 1987;12:302–306.
114. Gorkisch K, Boese-Landgraf J, Vaubel E. Treatment and prevention of amputation neuromas in hand surgery. Plast Reconstr Surg 1984;73:293–296.
115. Kon M, Bloem JJ. The treatment of amputation neuromas in fingers with a centrocentral nerve union. Ann Plast Surg 1987;18:506–510.
116. Barbera J, Albert-Pamblo R. Centrocentral anastomosis of the proximal nerve stump in the treatment of painful amputation neuromas of major nerves. J Neurosurg 1993;79:331–334.
117. Lidor C, Hall RL, Nunley JA. Centrocentral anastomosis with autologous nerve graft treatment of foot and ankle neuromas. Foot Ankle Int 1996;17:85–88.
P.865

118. Viterbo F, Trindade JC, Hoshino K, et al. End-to-side neu-rorrhaphy with removal of the epineural sheath. An experimental study in rats. Plast Reconstr Surg 1994;94:1038.
119. Viterbo F, Trindade JCS, Hoshino K, et al. Latero-terminal neurorrhaphy without removal of the epineural sheath. Experimental study in rats. Rev Paul Med 1992;110:267–275.
120. Edds MN. Prevention of nerve regeneration and neuroma formation by caps of synthetic resin. J Neurosurg 1945;2:507–509.
121. Garrity RW. The use of plastic and rubber tubing, in the management of irreparable nerve injuries. Surg Form 1955;6:517.
122. Pinto V, Junqueira L. A comparative study of the methods for the prevention of amputation neuromas. Surg Gynecol Obstet 1954;99:492.
123. Coburn DF. Painful stumps and their treatment. US Nav Med Bull 1945;44:1194.
124. Spurling RG. The use of tantalum wire and foil in the repair of peripheral nerves. Surg Clin North Am 1943;23:1491.
125. White JC, Hamlin H. New uses of tantalum in nerve suture. Control of neuroma formation. J Neurosurg 1945;2:402.
126. Chiu DT, Wu J. Treatment of painful neuromas: a case report. Ann Plast Surg 2000;44:340–342.
127. Hodor L, Barkal K, Hatch-Fox LD. Cryogenic denervation of the intermetatarsal space neuroma. J Foot Ankle Surg 1997;36:311–314.
128. Davies E, Pounder D, Mansour S, et al. Cryosurgery for chronic injuries of the cutaneous nerve in the upper limb. J Bone Joint Surg Br 2000;82:413–415.
129. Kuzbari R, Liegl C, Neumayer C, et al. Effect of CO2 milliwatt laser on neuroma formation in rats. Lasers Surg Med 1996;18:81–85.
130. Konig HJ, Themann H, Hiller U, et al. Prevention of neuroma formation by neodym YAG laser: experimental observations. Acta Neurochir (Wien) 1987;87:63–69.
131. McGrath MH. Local steroid therapy in the hand. J Hand Surg [Am] 1984;9:915–921.
132. Smith JR, Gomez NH. Local injection therapy of neuromata of the hand with triamcinolone acetonide. A preliminary study of twenty-two patients. J Bone Joint Surg Am 1970;52:71–83.
133. Reddy PD, Zelicof SB, Ruotolo C, et al. Interdigital neuroma. Local cutaneous changes after corticosteroid injection. Clin Orthop 1995;317:185–187.
134. Basadonna PT, Rucco V, Gasparani D, et al. Plantar fat pad atrophy after corticosteroid injection for an interdigital neuroma: a case report. Am J Phys Med Rehabil 1999;78:283–285.
135. Brown H, Flynn JE. Abdominal pedicle flap for hand neuromas and entrapped nerves. J Bone Joint Surg Am 1973;5:575.
136. Holmberg J, Ekerot L. Post-traumatic neuralgia in the upper extremity treated with extraneural scar excision and flap cover. J Hand Surg [Br] 1993;188:111.
137. Millesi H, Schobel G, Bochdansky T. Subpectoral gliding tissue flap. Plast Reconstr Surg 1994;93:842.
138. Millesi H, Zoch G, Roth T. The gliding apparatus of peripheral nerve and its clinical significance. Ann Hand Surg 1990;9:87.
139. Jones NF. Treatment of chronic pain by “wrapping” intact nerves with pedicle and free flaps. Hand Clin 1996;12:765–772.
140. Foucher G, Sammut D, Greant P, et al. Indications and results of skin flaps in painful digital neuroma. J Hand Surg [Br] 1991;16:25–29.
141. Godetta GA, Rayan GM. Medial triceps flap coverage for an ulnar neuroma. Orthop Rev 1993;22:603–606.
142. Rose J, Belsky MR, Millender LH, et al. Intrinsic muscle flaps: the treatment of painful neuromas in continuity. J Hand Surg [Am] 1996;21:671–674.
143. Lamey MD. Grafting of digital nerves. In: Blair WF, ed. Techniques in hand surgery. Baltimore: Williams & Wilkins, 1996:767–774.
144. Mayo-Robinson AW. Nerve grafting as a means of restoring function in limbs paralyzed by gunshot or other injuries. BMJ 1917;1:117–118.
145. Millesi H, Meissl G, Berger A. Further experience with interfascicular grafting of the median, ulnar and radial nerves. J Bone Joint Surg Am 1976;58:209–218.
146. Taylor GI, Ham F. The free vascularized nerve graft. Plast Reconstr Surg 1976;57:413–426.
147. Tarlov IM, Epstein JA. Nerve grafts: the importance of an adequate blood supply. J Neurosurg 1945;2:49–71.
148. St. Clair Strange FG. An operation for nerve pedicle grafting: preliminary communication. Br J Surg 1947;34:423–425.
149. Watchmaker GP, Gumico CA, Crandal RE, et al. Fascicular topography of the median nerve. J Hand Surg [Am] 1991;16:53–59.
150. Brushart TM. Nerve repair and grafting. In: Green DP, Hotchkiss RN, Pederson WC, eds. Green’s operative hand surgery, vol. 2, 4th ed. New York: Churchill Livingstone, 1999:1381–1403.
151. Mackinnon SE, Glickman LT, Dawgum A. A technique for the treatment of neuroma-in-continuity. J Reconstr Microsurg 1992;8:379–383.