When sensory nerve fibers are severed, the proximal axons remain in continuity with cell bodies in the dorsal root ganglion. These axons sprout and seek Schwann-cell guides. Schwann cells are believed to upregulate expression of neurotrophic factors, which induce axonal growth.^2,3,4,5 If the perineurium of the injured nerve has remained intact, as in the case of crush or certain thermal injuries, the sprouts may successfully reinnervate the target tissue. Successful regeneration may also occur when the transected ends of the nerve are surgically reapproximated (neurorrhaphy). However, when Schwann-cell guides are not present, the axon sprouts are unable to reach the target tissue and randomly double back on themselves. This disordered process of growth ultimately results in a densely packed cluster of nerve sprouts known as a neuroma.

Weir Mitchell⁶ brought attention to the problem of painful nerve injury after caring for wounded soldiers during the American Civil War. A century later, Denny-Brown and Kirk² presented one of the first studies demonstrating that axotomy can induce behavioral signs of pain in an animal model. More recent studies have suggested that axotomy of a major nerve, by itself, may induce hyperalgesia in animals. Although nontraumatic neuropathies may induce pain in diverse ways, the single nerve lesion offers a useful model through which to understand the mechanisms of neuropathic pain.

At least four pathophysiologic mechanisms appear to play a role in nerve injury pain.

Ectopic generation of action potentials: Although normally silent, nociceptive afferents may become spontaneously active following nerve injury, producing action-potential activity in the absence of a stimulus.³ This activity may be experienced as spontaneous pain. In addition to abnormal signaling in the nerve itself, the activity may sensitize central neurons, such that inputs from nonnociceptive, tactile afferents produce pain (allodynia).⁷

Ectopic excitability: Uninjured nerve trunks are minimally sensitive to mechanical stimuli. Gentle percussion over a nerve is not painful. Following injury, however, regenerating fibers may abnormally respond to mild, mechanical stimuli. Such ectopic mechanical excitability gives rise to Tinel sign, an electrical sensation in the nerve’s original target distribution, elicited by mechanical stimulation at the location of regenerating axons. Furthermore, ectopic excitability to mechanical stimuli may be accompanied by chemical sensitization. For example, injured nociceptive axons may become abnormally sensitive to catecholamines. As a result, the physiologic release of norepinephrine from sympathetic terminals may induce pain (sympathetically maintained pain [SMP]).^4,5

Nervi nervorum: Nerves themselves appear to be innervated by nociceptive fibers. These nervi nervorum fibers may be sensitized to mechanical stimuli following nerve injury. Such a mechanism may explain, for example, the local tenderness and
mechanical hyperalgesia of the ulnar nerve when it is entrapped at the elbow or the local tenderness of nerves entrapped by scar tissue.⁸

Ephaptic conduction: Under normal conditions, signals in adjacent afferent nerve fibers are insulated from each other. Activity in an injured nerve fiber may cross to a nearby fiber, through a direct electrical connection between the two. During such ephaptic transmission, or cross-talk, a nonnoxious sensory stimulus may evoke activity in nociceptive fibers and thereby cause pain.

Some of the mechanisms of ectopic generation of action potentials and ectopic excitability have been described. When an axon is severed, the axonal transport of sodium channels and other ion channels from the neuronal cell body to the sensory terminal is interrupted. As a result, channels may be expressed ectopically in the neuroma formed at the nerve injury site. Nociceptive fibers in the neuroma thereby become sensitive to normally nonpainful stimuli, producing pain when these stimuli are present.^9,10 In addition, nerve injury may lead to profound changes in gene expression, promoting ectopic excitability.¹¹

Other mechanisms may also contribute to neuropathic pain. Although inflammatory and neuropathic pain syndromes are traditionally considered separately, immunologic studies have implicated several pathways through which inflammatory responses may alter nociceptive processing, resulting in neuropathic pain.¹² Evidence supports what may be termed the wallerian degeneration hypothesis.¹³ According to this hypothesis, uninjured nociceptors that are adjacent to nerves undergoing wallerian degeneration may become spontaneously active and develop sensitivity to catecholamines, resulting in spontaneous pain and SMP. To the extent that neuropathic pain results from these mechanisms, peripheral neurectomy may be expected to worsen pain because the nerve undergoes wallerian degeneration distal to the site of neurectomy.

The concept of surgery to remove a neuroma as a treatment of nerve injury pain is a flawed one. Neuromas arise from nerve fibers proximal to a region of transection or severe injury, which remain in continuity with their cell bodies in the dorsal root ganglia. Cutting the nerve at a location that is proximal to the site of nerve injury to remove the neuroma results in the formation of a new neuroma at the proximal location. Surgery in which neuromas are “removed” should therefore be termed neuroma relocation surgery.

Not all neuromas are painful. The tissue milieu surrounding the nerve may determine whether a neuroma becomes painful or remains painless. The use of peripheral neurectomy as a treatment for painful neuromas is therefore predicated on the hope that relocation of the neuroma may convert it from a painful one to a painless one. For example, relocating a neuroma to a non-pressure-sensitive area may alleviate pain in some patients. Relocation of neuromas into muscle was first described in 1918. Since that time, the identification of anatomic locations appropriate for nerve relocation has improved outcomes substantially.^{14,15,16,17,18,19,20,21,22}

Thus, an important consideration in peripheral neurectomy is the role of location in the production of pain. If the location of nerve injury contributes to pain, then relocating the neuroma to a mechanically favorable area may be advantageous. However, to the extent that there is location-independent ectopic generation of action potentials in the neuroma, neuroma relocation surgery will fail. Additionally, some investigators have argued that central mechanisms may account for pain in many nerve injuries.²³ In this circumstance, neuroma relocation surgery may also fail. However, our observations as well as those of other experienced clinicians suggest that in patients with nerve injury pain, where anesthetic block of the injured nerve relieves pain, peripheral neurectomy may provide significant pain relief.

Clinical scenarios favoring peripheral neurectomy may be broadly divided into two circumstances: neuropathic pain resulting from nerve injury and nociceptive pain from a diseased tissue other than nerve. Nerve injury pain is characterized by numbness, burning, and allodynia. Tinel sign may be present at the site of a painful neuroma. Candidates for neuroma relocation surgery should respond to local anesthetic blockade.

Successful anesthetic blockade is an important prerequisite for effective neuroma relocation surgery. If blockade of the putative, pain-generating neuroma fails to relieve pain nearly entirely, the rationale for neuroma relocation surgery is precarious. The decision to operate may be made with more confidence if more than one block is done. Once candidate nerves are identified, local anesthetic blocks indicate the level of benefit that can be obtained following nerve ablation. A successfully applied block should induce anesthesia in the distribution of each target nerve, but not beyond, to indicate the specificity of the blockade. Injection of saline or injections away from the nerve may enhance blockade specificity by identifying nonspecific responses (e.g., placebo responses).

Findings associated with complex regional pain syndrome (CRPS), such as edema, hyperalgesia, and trophic changes, also suggest that neurectomy will not alleviate pain. Notably in such cases, peripheral nerve blockade typically produces little relief. Patients should be additionally assessed for hyperalgesia to cooling stimuli. This finding is suggestive of SMP, discussed in the following text. Finally, local tenderness in combination with Tinel sign suggests nerve entrapment. In this instance, nerve decompression would be indicated rather than neurectomy. It is important to note that even subtle entrapments without significant motor or sensory loss may induce severe pain.

Even after one has identified a specific nerve as the pain generator, and one has determined that there are no contraindications to neuroma relocation surgery, a wait-and-see approach may remain most appropriate. For example, where injury is relatively recent and the nerve relatively minor, one may choose to observe. The pain may resolve spontaneously. Where there is only partial nerve injury with remaining function, neurectomy may sacrifice function without any assurance that the new neuroma will be less painful than the old one. In this circumstance, nerve repair should be considered before performing neuroma relocation surgery.

Nerve repair has the potential to relocate neuromas with the advantage of restoring neurologic function. The clinician sometimes faces the ironic situation that to repair an injured nerve, a normal nerve (e.g., sural nerve) may be sacrificed to provide donor grafts. In effect, nerve repair is, in a sense, still a neuroma relocation operation—the neuroma is relocated to the donor nerve. Repair of an injured nerve, when feasible, is generally preferable to permanent transection. By contrast, if a nerve has already been completely severed, the risk of relocating the neuroma is low. Thus, in a case of a well-defined neuroma, when an anesthetic block relieves pain, surgical neurectomy may be the preferred first line of surgical therapy.

Careful analysis may lead to rewarding outcomes. The following case presents the history, preoperative evaluation, and treatment of a patient with neuropathic pain.

A 44-year-old woman presented with a chief complaint of right vaginal pain. This problem had been present for 3 years and originated with an excisional biopsy of a right-sided vaginal ulcer near the introitus. The pain was always present, but especially disturbing were lightning attacks of pain that occurred unpredictably several times a day. Examination disclosed a
subtle sensory loss in the right vulvar area. Medication trials were minimally helpful. An anesthetic block of the right pudendal nerve led to 50% pain relief. A combined ilioinguinal and genitofemoral nerve block also led to 50% pain relief. A local anesthetic block of all three nerves together led to 100% pain relief. As treatment, the right pudendal nerve was severed distal to the sacral spinous ligament through a perivulvar approach, and the patient, predictably, had 50% of her pain relieved. At a separate surgery, the right ilioinguinal and genitofemoral nerves were severed through a retroperitoneal approach. At 3-year follow-up, the patient had complete relief. There were no adverse sequelae.

In this case, both lumbosacral neural segments provided innervation to the painful neuromas. Failure to appreciate this would have led to a less than satisfactory result. This case underscores the need for complete blockade of pain during the application of local anesthetic to ensure that all involved nerves are identified.

Once peripheral neurectomy has been selected as the treatment of choice, the primary surgical issue is where to relocate the neuroma. Troublesome neuromas typically are in areas near joints, scars, and structures that may tether the nerve. The idea of surgery is to relocate the neuroma to a new location where tethering does not occur.¹⁷ Nerves may be cut back to locations such that the ends can be placed in healthy, well-vascularized muscle. Some have also advocated that neuromas be placed in holes in bone. Placement of a cut nerve into these environments does not change the fact that the cut ends of the nerve will sprout and that a neuroma will form. However, with relocation of the neuroma into muscle or bone, chances are reduced that the new neuroma will be subject to the tension and shearing forces likely to play a role in pain generation.

Alternatives to neuroma relocation exist. The nerves may be cauterized, frozen, burned, or injected with toxic chemicals. These options have been reported to be successful, but their advantages over surgical neurectomy have not been demonstrated. A surgical procedure where the neurectomy is done sharply, with limited damage to the surrounding environment of the nerve, is most appealing from a mechanistic perspective. Damage to the surrounding tissues, such as necrosis due to phenol injection, may create a new focus for pain generation.²⁴

In cases of stump pain, it is worthwhile to examine the patient for tender neuromas. A prosthetic device, for example, may apply pressure to the neuroma, causing pain. Surgical relocation of neuromas to more proximal or protected locations, often in conjunction with nerve wrapping, may provide significant benefit in these cases.^25,26

Chest trauma or thoracotomy may damage intercostal nerves. Shoulder trauma and axillary node dissection may damage the intercostobrachial nerve. Motor deficits associated with intercostal and intercostobrachial neurectomy are clinically insignificant. Hence, neurectomy is usually without significant drawbacks. This procedure can be safely accomplished through video-assisted thoracoscopy as well as through an open procedure.²⁷

Injuries to the pudendal, ilioinguinal, iliohypogastric, and genitofemoral nerves may result in severe pain. These injuries are often due to abdominal and pelvic surgery, episiotomy, hernia repair, entrapment, or blunt trauma. For example, the Pfannenstiel transverse incision may injure the ilioinguinal/iliohypogastric nerves. Groin pain from inguinal herniorrhaphy is not uncommon. Stulz and Pfeiffer²⁸ reported relief of pain with neurectomy in 70% (16 of 23) of patients with ilioinguinal and iliohypogastric neuralgia as a complication of prior surgery. Starling and Harms²⁹ reported similar rates of success: 89% (17 of 19) for ilioinguinal neuralgia and 71% (12 of 17) for genitofemoral neuralgia. In the largest series to date, Amid³⁰ reported 95% improvement in pain among 225 patients. Our own experience supports the use of neurectomy, but the incidence of long-term favorable outcomes is much more modest.³¹ Recently, Chen and colleagues³² reported that patients who underwent a laparoscopic retroperitoneal triple neurectomy for inguinal pain over open and standard procedures for inguinal herniorrhaphy showed superior outcomes in terms of postoperative pain scores and recovery time.

Entrapment or injury of the lateral femoral cutaneous nerve, meralgia paresthetica, may result in pain and dysesthesia in the anterolateral thigh. In cases where the diagnosis is unclear, local anesthetic blockade may be helpful. Transection should ordinarily be considered as a backup procedure when decompression is not feasible and nonsurgical approaches, such as weight loss, have failed.³³ A study comparing neurolysis to neurectomy for the treatment of meralgia paresthetica found that only 60% of patients who underwent neurolysis reported being pain free compared to 87.5% of patients treated with neurectomy.³⁴

Entrapment of the saphenous nerve in the subsartorial canal³⁵ may occur with or without a history of trauma. Damage to the saphenous nerve may occur when the saphenous vein is harvested during revascularization procedures.³⁶ Risk factors associated with the development of saphenous neuralgia after saphenous vein harvest include younger age, female sex, diabetes mellitus, higher body mass index, distal-to-proximal dissection of the saphenous vein, and closure of the leg wound in two layers.³⁷ The condition is associated with pain (with or without numbness) along the anterior and medial leg and the dorsum of the foot. Proximal neurectomy may be used if the nerve has been directly injured. Otherwise, in our experience, the nerve should be decompressed as the first-line treatment.

This condition involves compression of the digital nerve, typically between the third and fourth tarsal bones. The compression leads to a swelling of the nerve, which is mistakenly called a neuroma. Patients present with pain in this region worsened by wearing shoes and walking. If conservative measures (e.g., orthotics) fail, neurectomy may be offered. Indeed, this operation is a common procedure for this condition. Johnson et al.³⁸ reported relief of pain in 67% (22 of 33) of patients, with 6 years average follow-up, following excision of the plantar interdigital neuroma. Others have reported surgical success rates of up to 90%.^39,40 A recent consecutive cohort study reported that ultrasound-guided radiofrequency ablation of
Morton’s neuroma provided pain relief in 25 out of 30 patients. This procedure can be completed on an outpatient basis and is less invasive than traditional neurectomy.⁴¹

What are the predicted results of neurectomy for nerve injury pain? The question is difficult to answer because the patients undergoing this treatment are heterogeneous. In addition, measurement of patient outcome varies greatly among studies with regard to methodologic rigor, length of follow-up, and technique. As suggested by the studies cited earlier, success rates vary from modest to high.

Burchiel and colleagues⁴² have taken a systematic approach to the treatment of nerve injury pain, moving the field toward a definition of the indications for neuroma surgery. In their study, 42 patients with nerve injury pain were divided into four treatment groups:

Surgical success (rated as a greater than 50% subjective improvement in pain levels, subjectively rated pain relief as “good” or “excellent,” and no postoperative narcotic usage) varied between the groups. In the 40 patients who received postoperative follow-up care over 2 to 32 months (average of 11 months), 16 (40%) met these criteria. By group, successful pain relief was accomplished in 44% (8 of 18) of group 1, 40% (4 of 10) of group 2, 0% (0 of 5) of group 3, and 57% (4 of 7) of group 4.

After obtaining these results, Burchiel et al.⁴² attempted to determine retrospectively the extent to which indicators of nerve injury predicted surgical success. Such indicators included Tinel sign, hyperalgesia, a “discrete nerve syndrome,” litigation, and prior procedures. Some predictors showed promise. For example, a discrete nerve syndrome, defined as a condition in which a single nerve could account for all the neurologic findings and pain distribution, tended to predict success. However, none of the relationships between preoperative diagnostic variables and treatment success achieved statistical significance at the P < .05 level. Another prospective study found that employment status, duration of pain, CRPS, smoking, and improvement with nerve block were all prognostic factors for surgical management of neuroma pain.⁴³

One can only speculate why the results of this series differ substantially from those of other series. Perhaps patient selection accounts for differences, yet Burchiel and colleagues⁴² appeared to discriminate patients at high risk for failure. Surgical technique could also play a role, but there is no evidence on which to base such a statement. In some cases, neuromas are innervated by more than one nerve. For example, proximal resection of the superficial radial nerve to treat dorsoradial wrist neuromas often relieves pain only temporarily. Further inspection reveals that the lateral antebrachial cutaneous nerve may also innervate these neuromas, and thus, success may require sectioning this nerve as well.⁴⁴

Another reason for failure in neuroma relocation surgery may relate to the discovery that the distal side of a severed nerve may also form a neuroma at the site of injury. Plexus formation distal to the neurectomy may allow intact nerve fibers from other nerves to sprout in retrograde fashion to innervate this distal site. This retrograde sprouting may create a potentially painful neuroma on the “wrong” side.⁴⁵ Neuroma relocation surgery should perhaps attend to neuroma formation on both sides of a severed nerve.

The medial branches of dorsal rami innervate the paraspinal muscles, the interspinous ligament, and the zygapophyseal (facet) joints. Pain associated with movement of the lower back, which is relieved by rest and is not attributable to other spine pathology, may be relieved through bilateral, percutaneous radiofrequency or chemical ablation of these branches in the lumbar spine.^46,47 A success rate of 85% with mean duration of relief of 10.5 months was reported by Schofferman and Kine⁴⁸ through this procedure. In addition, neck pain associated with whiplash injury may benefit from a similar procedure in the cervical spine.⁴⁹ The mechanism of pain relief is thought to relate to denervation of the facet joint. Because the dorsal ramus innervates several structures, other mechanisms are possible. Evidence is indeterminate for similar pain of thoracic origin.

Many experts suggest diagnostic anesthetic blocks of the facets prior to a facet denervation procedure. Good results are expected only in patients who get excellent benefit from the blocks. Patients who have axial pain in addition to radicular symptoms tend not to benefit from facet denervation procedures alone. Among preexisting symptoms, only paraspinal tenderness has been shown to predict treatment success, whereas increased pain with facet loading maneuvers predicts less favorable outcomes.^50,51

The procedure is performed percutaneously with radiofrequency heat lesions. The advantages of this procedure are that it can be done on an outpatient basis and morbidity is low. The disadvantage is that the procedure often confers only temporary relief or often no relief at all regardless of the temporary effects of diagnostic facet blocks. In addition to continuous, hightemperature radiofrequency medial branch ablation, pulsed radiofrequency, cryodenervation, and phenol neurolysis have also been used to provide intermediate to long-term pain relief.

In patients with no prior spine surgery, these procedures have been reported to provide initial relief for 60% to 70% of patients.^36,49,50,52 Rates are reported to be considerably lower, 20% to 50%, for patients with prior spine surgery.^36,52,53 However, a history of spine surgery is associated with treatment failure not only for radiofrequency denervation but other interventions as well, including epidural steroid injection and open surgery. Our view is that this is a low-morbidity procedure that can provide effective, albeit impermanent, pain relief for patients with axial spine pain. Recurrence of pain following denervation can be treated with repeated neurotomy with comparable efficacy. See Chapter 102 for a detailed discussion of neurolytic blocks, including radiofrequency neurolysis.

Denervation procedures aim to eliminate pain arising from degenerative processes in the joint while preserving functions that may be lost after other forms of joint surgery. Buck-Gramcko⁵⁴ reported retention of wrist mobility with substantial reduction of pain in 69% (135 of 195) of patients following wrist denervation surgery. Wilhelm⁵⁵ reported success in 90% of patients with tennis elbow treated by denervation. Dellon et al.⁵⁶ reported satisfaction in 86% (60 of 70) of patients following partial denervation surgery for persistent, postoperative knee pain. Pulsed radiofrequency denervation has also been used to provide relief although generally with shorter effect when compared to conventional radiofrequency ablation and limited evidence to support the efficacy of this approach.⁵⁷

Neurectomy of the superior hypogastric (presacral) plexus has been advocated as a treatment for medically refractory pelvic pain. In 1948, Ingersoll and Meigs⁵⁸ reported complete relief of primary dysmenorrhea in 81% (72 of 89) of women treated with neurectomy. More recently, with the development of more effective analgesics, ablative approaches to pelvic pain have been largely limited to patients with secondary dysmenorrhea associated with endometriosis. Nezhat et al.⁵⁹ reported at least 50% relief from pain in 70% to 85% of patients with various stages of endometriosis, with 1-year follow-up, following presacral neurectomy combined with excision and vaporization of endometriotic lesions. Debate over the proper role of presacral neurectomy has been ongoing for well over 50 years. Introduction of the biopsychosocial model of chronic pain has potential to spare many women from surgery.

Peripheral neurectomy is infrequently used in the treatment of pain due to cancer. This is due to the availability of alternative strategies with low morbidity and higher success rates, such as spinal opiates or percutaneous radiofrequency cordotomy.^60,61 Transecting a peripheral nerve may fail to relieve pain because of overlapping receptive fields of adjacent nerves or central plasticity. Sectioning major peripheral nerves results not only in numbness but also in unacceptable motor loss. Peripheral neurectomies are rarely indicated in the extremities. However, localized chest or abdominal wall pain can successfully be treated with intercostal neurectomies. Alcohol injection, cryoprobe, or radiofrequency lesions provide similar success to open surgery and lowered morbidity for the treatment of cancer pain.

Nerve entrapment syndromes result from pressure applied directly to a nerve, causing pain, paresthesias, or weakness in the sensory distribution of the nerve.^44,62,63 Entrapment commonly develops where peripheral nerves traverse confined anatomic spaces, rest in superficial locations or in proximity to joints, or become tethered to adjacent tissues. Structural factors such as anomalous nerves, cervical ribs, muscles, or connective tissue bands also may contribute. Repetitive motion, nerve traction due to joint position, chronic vibration, and high force constitute physical stressors that increase the risk of symptomatic entrapment.

Although the degree of compression required to cause nerve injury may vary, axonal degeneration follows pressure in a dose-dependent relationship.⁶⁴ Symptoms can arise following a few significant events or after a longer period of repetitive mild insults. The time course and force of injury are just two of the prognostic considerations. The length of the affected nerve region as well as the presence and severity of nerve ischemia are important factors in the development of symptomatic nerve entrapment. Finally, nerves become more vulnerable to injury in the presence of concurrent systemic metabolic disease, and nerves lose their regenerative capacity with age.

Several pathophysiologic mechanisms have been proposed to explain pain associated with peripheral nerve entrapment at the cellular level. Evidence for these mechanisms arises from studies of animal models.^64,65,66 Following local nerve compression, internodes along myelinated fibers distort in shape. Demyelination appears earliest in the segment nearest to the point of compression.^67,68 Eventually, segmental demyelination leads to diffuse demyelination and, ultimately, axonal degradation. Compressive injury usually affects larger, more peripherally located myelinated nerves as opposed to smaller or unmyelinated fibers.

Nerve ischemia may also play an important role in nerve entrapment pain syndromes. Focal pressures of 20 to 30 mm Hg may impede venous blood flow, whereas higher pressures may reduce arterial supply.^69,70 Within 4 hours of extraneural compression, increasing permeability of the blood-nerve barrier leads to development of subperineurial edema.^71,72,73,74 As there is no lymphatic drainage of the endoneurial space, sustained intraneural pressure elevations persist for at least 24 hours after compressive forces are removed. Following such injury, reactive inflammation, fibrin deposition, and proliferation of endoneurial fibroblasts and capillary endothelial cells lead to intraneural fibrosis of perineurial and epineurial tissues.

Ischemia does not appear to play a role in the initial demyelination associated with acute compressive injury.^75,76 Rather, nodes of Ranvier adjacent to the point of compression are outwardly displaced and disrupt the myelin sheath. Segmental demyelination follows, stemming from these sites of myelin invagination.^77,78 Disruption of anterograde axoplasmic flow of vital nutrient proteins reduces terminal membrane excitability. Long-standing ischemia thereby results in replacement of funicular contents with fibrotic tissue, producing derangements in electrical conductivity.⁷⁹

Several endocrine and rheumatologic diseases show an association with increased risk for nerve entrapment. Approximately 15% of upper extremity nerve entrapment patients have diabetes. This may result from an increased association of peripheral neuropathies with entrapment. Both hypo- and hyperthyroidism have been shown to pose an increased risk for nerve entrapment, which is thought to result from glycogen deposition in Schwann cells.⁸⁰ Up to 30% of acromegalics are diagnosed with nerve entrapment syndromes, which often resolve with treatment of the underlying acromegaly.^81,82 Obesity and pregnancy have well-documented associations with entrapment syndromes, with as many as two-thirds of pregnant women experiencing temporary symptoms.⁸³ Rheumatoid arthritis is also thought to contribute to entrapment in anatomic locations where synovial overgrowth can produce compression of a nerve. There is an estimated 45% incidence of entrapment neuropathy in rheumatoid arthritis patients.^84,85 Amyloidosis, carcinomatosis, gout, mucopolysaccharide storage diseases, and polymyalgia rheumatica are all suspected to pose an increased risk for exacerbation of nerve entrapment.

Clinical findings: Patients suffering from nerve entrapment syndromes present with pain, paresthesias, or motor weakness along a specific nerve distribution. Nocturnal pain, either sharp or burning, classically develops before the onset of daytime or persistent symptoms. Muscular changes, such as atrophy or
fasciculation, as well as sensory alterations, such as altered two-point discrimination or temperature sensation, can occur in advanced cases. Specific symptoms result from the location and severity of compression.

On physical examination, Tinel sign may be present. Tinel sign is an electric radiating sensation in the distribution of the nerve, produced by percussion over the nerve. Tinel sign represents ectopic excitability, a hallmark of nerve entrapment, and may be accompanied by local tenderness. Provocative maneuvers may reproduce or exacerbate symptoms and have significantly positive predictive value in the diagnosis of entrapment syndromes.

In many cases, diagnosis of a specific entrapment syndrome may be difficult to make. Compensation of one muscle group by another may create uncertainty in identifying etiology of the entrapment. Furthermore, entrapment syndromes may be accompanied by concurrent radiculopathy, producing synergistic pain, thereby adding to the diagnostic challenge.⁸⁶ In addition, pain associated with entrapment syndromes may not be well localized. For example, radial nerve entrapment may present with global arm pain. Finally, results of electrodiagnostic studies may be normal in some cases, further increasing the diagnostic dilemma.

Electrodiagnostic studies: Physical findings may be supported by electrodiagnostic studies. Nerve conduction studies (NCS) measure action potentials along axons, whereas electromyography studies (EMG) measure muscle fiber activity. In sensory NCS, a stimulus is applied to the nerve and a sensory nerve action potential (SNAP) is measured at various distal points along the nerve. Alterations in axon number, diameter, myelination, or temperature all affect the magnitude and temporal profile of the SNAP. In motor NCS, a stimulus is applied to the nerve and a compound motor action potential (CMAP) is measured from the innervated muscle. Changes in conduction velocity, amplitude of CMAP, and distal motor latency suggest that conduction inhibition exists between the point of initial stimulation and the site of recording. Needle EMG examines the spontaneous electrical activity of individual muscle fibers resulting from denervation, reinnervation, or acute muscle injury.

These studies aid in the differentiation of peripheral nerve entrapments from brachial plexopathy and radiculopathy by defining the nature and extent of neurologic dysfunction. For example, although NCS cannot be performed in structures like the brachial plexus due to an inability to record and stimulate across the region of compression, a plexus lesion would be expected to affect the function of multiple peripheral nerves. In entrapment neuropathy, large myelinated sensory fibers are typically affected before motor fibers. Hence, NCS are most often effective earlier in the disease than motor conduction studies (ulnar compression at the elbow is a notable exception to this rule). Entrapment is represented by abnormal recordings of evoked SNAPs along a short nerve segment. Prolonged latency or reduced motor conduction velocity is also indicative of injury, although they are less sensitive measures. Fibrillation seen on EMG occurs as a result of wallerian degeneration of distal nerve segments and suggests advanced entrapment. In general, a combination of studies and measurement points yields the best information with which to form a diagnosis.

Comparison among the nerves of interest and a nearby unaffected nerve is often more accurate than comparison of a nerve with normal reference values. Cooler temperature, increasing age, and patient size all serve to prolong sensory latencies, confounding reference values. As a result, and to avoid such confounding factors, it is common to perform sensory latency comparisons at as many as three sites (medianulnar midpalmar, median-ulnar ring finger, and median-radial thumb) in the diagnosis of carpal tunnel syndrome (CTS).⁸⁷ Although the greater number of studies increases the likelihood of false-positive results,⁸⁸ summation of the latencies of these three studies has proven to be the most sensitive and specific test for CTS.⁸⁹

Following entrapment release, it is notable that clinical improvement may not correlate reliably with electrodiagnostic studies. However, in patients with persistent symptoms following treatment, comparison with pretreatment studies may be helpful. If earlier studies are unavailable, a series of postoperative studies over several months should be obtained.

Imaging studies: Developments in ultrasonography and magnetic resonance imaging (MRI) have resulted in a potentially expanded role for imaging in the diagnosis of entrapment neuropathies. Ultrasonography now offers the highest available resolution for visualizing the location of entrapment (Fig. 103.1). However, ultrasound is unable to show pathologic changes within nerves. In contrast, signal and configuration characteristics seen in chronically compressed or tethered peripheral nerves have been well characterized by MRI (Fig. 103.2). On MRI, indicators for nerve abnormality include focal enlargement, T2 hyperintensity, and an indistinguishable or nonuniform fascicular pattern. In addition, MRI simultaneously visualizes muscular alterations due to atrophy and neurogenic edema.

Specific MRI findings have been correlated with clinical findings, electrodiagnostic findings, and postoperative outcomes for several nerve entrapment syndromes.^90,91,92,93 With current technology, MRI may be a useful adjunct in the diagnosis of entrapment. Specific indications for MRI may include ulnar entrapment, entrapment in the presence of superimposed neuropathy, entrapment where other tests are equivocal, and postoperative evaluation of entrapment.⁹⁴ The value of MRI imaging in the setting of uncommon entrapments is less clear.

Substantial experimental evidence supports three treatments of peripheral nerve entrapment: splinting, local corticosteroid injection, and surgical release. Splinting reduces the offending stimulus through postural correction and reduction of repetitive injury, thereby decreasing reactive inflammation. Local corticosteroid injections have been shown to be helpful in mild cases. Surgical release of anatomic compression is the mainstay of treatment in more advanced cases of peripheral nerve entrapment.⁹⁵

In general, operative release of entrapment can be quite rewarding, as the risks of surgery are typically low and pain improvement is often dramatic. Avoidance of additional trauma to the compressed nerve is of great importance to surgical outcome. Occasionally, unexpected sources of entrapment may be found during surgical exposure, such as cysts, soft tissue masses, and bony prominences. Choices of surgical approach and operative technique are based largely on individual surgeon preferences.⁹⁶ Rates of success and complications vary with the location of entrapment and the etiology of pain.