Structure of Peripheral Nerves

The peripheral nerve is a cable-like structure containing bundles of both unmyelinated and myelinated fibers and their supporting elements. The unmyelinated axons are surrounded only by the plasma membrane of a Schwann cell. The myelinated axons are engulfed by a Schwann cell that wraps around the axons multiple times, thereby insulating the axon with multiple layers of lipid-rich cell membrane. The myelinated axon is surrounded completely by myelin and Schwann cells except at regular gaps called the nodes of Ranvier, which measure approximately 1 µm in adults. The propagation of action potentials from one node of Ranvier to the next (saltatory conduction) is maintained by a thick myelin sheath with low capacitance and high resistance to electric current and by a high concentration of voltage-gated sodium channels at the nodes of Ranvier.

Pathological Processes Involving Peripheral Nerves

Despite the large number of causes for neuropathy, the pathological reactions of peripheral nerves to various insults remain limited. In general, these pathological processes can be divided into four main categories: (1) wallerian degeneration, which is the response to axonal interruption, (2) axonal degeneration or axonopathy, (3) primary neuronal (perikaryal) degeneration or neuronopathy, and (4) segmental demyelination or myelinopathy. The patient’s symptoms, the type and pattern of distribution of signs, and the characteristics of nerve conduction study abnormalities provide information about the underlying pathological changes.

Any type of mechanical injury that causes interruption of axons leads to wallerian degeneration—that is, distal degeneration of axons and their myelin sheaths. Immediately following injury, motor weakness and sensory loss occur in the distribution of the damaged nerve, as well as complete loss of voluntary activity (with a complete lesion) or a decrease in motor unit action potential (MUAP) recruitment (with a partial lesion) on needle electromyogram (EMG). However, the axons remain excitable distally, since distal conduction failure is not completed until 10 to 11 days later as the distal nerve trunk becomes progressively unexcitable. On nerve conduction studies, the amplitude of the compound muscle action potential (CMAP) begins to decline by the second day after injury and reaches its nadir by the fifth to sixth. For sensory axons, the loss of sensory nerve action potential (SNAP) is delayed by another 2 to 3 days; distal SNAP remains normal for 5 to 6 days and then decreases rapidly to reach its nadir by 10 to 11 days after injury (see Fig. 32B.9 in Chapter 32B). The temporal sequence of wallerian degeneration is length dependent, occurring earlier in shorter than in longer distal nerve stumps. Denervation potentials (fibrillation potentials) are typically seen on needle EMG in some affected muscles 10 to 14 days after injury and become full after 3 weeks from acute nerve injury. Axonal interruption initiates secondary morphological changes of the nerve cell body, termed chromatolysis, and the proximal axonal caliber becomes smaller. Regeneration from the proximal stump begins as early as 24 hours following transection but proceeds slowly at a maximal rate of 2 to 3 mm/day and is often incomplete. The quality of recovery depends on the degree of preservation of the Schwann cell/basal lamina tube and the nerve sheath and surrounding tissue, the distance of the site of injury from the cell body, and the patient’s age.

Axonal degeneration (or axonopathy), the most common pathological reaction of peripheral nerve, signifies distal axonal breakdown and is presumably caused by metabolic derangement within neurons or ischemia. Systemic metabolic disorders, toxin exposure, vasculitis, and some inherited neuropathies are the usual causes of axonal degeneration. The myelin sheath breaks down concomitantly with the axon in a process that starts at the most distal part of the nerve fiber and progresses toward the nerve cell body, hence the term dying-back or length-dependent polyneuropathy (Fig. 76.1). A similar sequence of events may occur simultaneously in centrally directed sensory axons, resulting in distal degeneration of rostral dorsal column fibers. The selective length-dependent vulnerability of distal axons could result from failure of the perikaryon to synthesize enzymes or structural proteins, from alterations in axonal transport, or from regional disturbances of energy metabolism. In some axonopathies, alterations in axon caliber, either axonal atrophy or axonal swelling, may precede distal axonal degeneration. Clinically, dying-back polyneuropathy presents with symmetrical distal loss of sensory and motor function in the lower extremities that extends proximally in a graded manner. The result is sensory loss in a stocking-like pattern, distal muscle weakness and atrophy, and loss of distal limb myotatic reflexes. Axonopathies result in low-amplitude SNAPs and CMAPs, but they affect distal latencies and conduction velocities only slightly. Needle EMG of distal muscles shows acute and/or chronic of denervation changes (see Chapter 32B). Because axonal regeneration proceeds at a maximal rate of 2 to 3 mm/day, recovery may be delayed and is often incomplete.

Fig. 76.1 Diagram of the main pathological events of distal axonal degeneration or axonopathy. Jagged lines indicate that either a toxin or a metabolic insult acts at multiple sites along motor and sensory axons in the peripheral nervous system (PNS) and central nervous system (CNS). Axonal degeneration begins at the most distal part of the nerve fiber and progresses proximally by the late stage. Recovery occurs by axonal regeneration but is impeded by astroglial proliferation in the CNS.

(Adapted from Schaumburg, H.H., Spencer, P.S., Thomas, P.K., 1983. Disorders of Peripheral Nerves. Davis, Philadelphia.)

Neuronopathy designates loss of nerve cell bodies with resultant degeneration of their entire peripheral and central axons. Either anterior horn or dorsal root ganglion cells may be affected. When anterior horn cells are affected, as in anterior poliomyelitis or motor neuron disease, focal weakness without sensory loss results. Sensory neuronopathy, or dorsal polyganglionopathy, means damage to dorsal root ganglion neurons that results in sensory ataxia, sensory loss, and diffuse areflexia (Fig. 76.2). A number of toxins such as organic mercury compounds, doxorubicin, and high-dose pyridoxine produce primary sensory neuronal degeneration. Immune-mediated inflammatory damage of dorsal root ganglion neurons occurs in paraneoplastic sensory neuronopathy and Sjögren syndrome. It is often difficult to distinguish between neuronopathies and axonopathies on clinical grounds alone. Once the pathological processes are no longer active, sensory deficits become fixed, and little or no recovery takes place.

Fig. 76.2 Diagram of the main pathological events of a sensory neuropathy or gangliopathy. A toxin (jagged lines) produces destruction of dorsal root ganglion (DRG) neurons, which results in degeneration of the peripheral-central axonal processes. Recovery is poor, as no axonal regeneration can take place. CNS, Central nervous system; PNS, peripheral nervous system.

(Adapted from Schaumburg, H.H., Spencer, P.S., Thomas, P.K., 1983. Disorders of Peripheral Nerves. Davis, Philadelphia.)

The term segmental demyelination (or myelinopathy) implies injury of either myelin sheaths or Schwann cells, resulting in breakdown of myelin with sparing of axons (Fig. 76.3). This occurs mechanically by acute nerve compression or chronic nerve entrapment and in immune-mediated demyelinating neuropathies and hereditary disorders of Schwann cell/myelin metabolism. Primary myelin damage may be produced experimentally by myelinotoxic agents such as diphtheria toxin or by acute nerve compression. Remyelination of demyelinated segments usually occurs within weeks. The newly formed remyelinated segments have thinner-than-normal myelin sheaths and internodes of shortened length. Repeated episodes of demyelination and remyelination produce proliferation of multiple layers of Schwann cells around the axon, termed an onion bulb. The physiological consequence of acquired demyelination, such as in inflammatory or compressive demyelination but not hereditary myelinopathies, is conduction block, which results in loss of the ability of the nerve action potential to reach the muscle, thereby producing weakness. Because the axon remains intact, there is little muscle atrophy. Relative sparing of temperature and pinprick sensation in many demyelinating polyneuropathies reflects preserved function of unmyelinated and small-diameter myelinated fibers. Early generalized loss of reflexes, disproportionately mild muscle atrophy in the presence of proximal and distal weakness, neuropathic tremor, and palpably enlarged nerves are all clinical clues that suggest demyelinating polyneuropathy. Nerve conduction studies or analysis of single teased nerve fiber preparations stained with osmium can confirm demyelination. Demyelination is present if motor and sensory nerve conduction velocities (NCVs) are reduced to less than 70% of the lower limits of normal, with relative preservation of response amplitudes. The presence of partial motor conduction block, temporal dispersion of CMAPs, and marked prolongation of distal motor and F-wave latencies are all features consistent with acquired demyelination (see Chapter 32B). Recovery depends on the extent of remyelination, and therefore clinical improvement may occur within days to weeks. It should be noted that in many demyelinating neuropathies, axonal degeneration may also coexist.

Fig. 76.3 Diagram of the main pathological events of primary segmental demyelination in immune-mediated inflammatory polyradiculoneuropathies. Attack by inflammatory cells causes patchy multifocal demyelination along nerve fibers but spares their axons. Recovery occurs by remyelination. Demyelinated segments become invested by several Schwann cells, resulting in a decrease in internodal length of those areas. CNS, Central nervous system; PNS, peripheral nervous system.

(Adapted from Schaumburg, H.H., Spencer, P.S., Thomas, P.K., 1983. Disorders of Peripheral Nerves. Davis, Philadelphia.)

Classification of Peripheral Nerve Disorders

There are several patterns of peripheral nerve disease (Box 76.1). Brachial, lumbar, and sacral plexopathy are discussed in Chapter 75, and radiculopathies are discussed in Chapter 74.

A mononeuropathy means focal involvement of a single nerve and implies a local process. Direct trauma, compression, entrapment, vascular lesions, and neoplastic compression or infiltration are the most common causes. Electrophysiological studies provide a more precise localization of the lesion than may be possible by clinical examination, can distinguish axonal loss from focal segmental demyelination, and sometimes may reveal a more widespread change indicating an underlying generalized polyneuropathy that has made the nerve susceptible to entrapment, as occurs in diabetes mellitus, hypothyroidism, acromegaly, alcoholism, and hereditary neuropathy with liability to pressure palsy (HNPP).

Multiple mononeuropathies, or mononeuropathy multiplex, signify simultaneous or sequential damage to multiple noncontiguous nerves. Confluent multiple mononeuropathies may give rise to motor weakness with sensory loss that can simulate a length-dependent peripheral polyneuropathy.

Polyneuropathy is most commonly characterized by symmetrical distal motor and/or sensory deficits that typically have a graded increase in severity distally and distal attenuation of reflexes. The sensory and motor deficits generally follow a length-dependent stocking-glove pattern. Most polyneuropathies are fairly symmetrical, but some are asymmetrical and may be confused with confluent mononeuropathy multiplex. A small number of polyneuropathies (e.g., that associated with acute intermittent porphyria [AIP]) can be predominantly proximal. It is helpful to determine the relative extent of sensory, motor, and autonomic neuron involvement, although most polyneuropathies produce mixed sensorimotor deficits and some degree of autonomic dysfunction.

Diagnostic Clues from the History

The symptoms of peripheral nerve disorders fall under the general headings of motor, sensory, or autonomic disturbances. The inquiry should seek both negative and positive symptoms. Negative motor symptoms are weakness, atrophy, and walking difficulties. Muscle cramps, fasciculations, myokymia, or tremor are positive manifestations of motor nerve dysfunction. In polyneuropathies, negative motor symptoms include early distal toe and ankle extensor weakness, resulting in tripping on rugs or uneven ground. However, a complaint of difficulty walking in itself does not distinguish muscle weakness from sensory, pyramidal, extrapyramidal, or cerebellar disturbance. If the fingers are weak, patients may complain of difficulty opening jars or turning a key in a lock.

Positive sensory symptoms include prickling, searing, burning, and tight band-like sensations. Paresthesias are unpleasant sensations arising spontaneously without apparent stimulus. The presence of spontaneously reported paresthesias is helpful in distinguishing acquired (>60% of patients) from inherited (≈17% of patients) polyneuropathies. Allodynia refers to the perception of nonpainful stimuli as painful, and hyperalgesia is painful hypersensitivity to noxious stimuli. Neuropathic pain, the extreme example of a positive symptom, is a cardinal feature of many neuropathies. Neuropathic pain often has a deep, burning, or drawing character that may be associated with jabbing or shooting pains that typically increases at night or during periods of rest.

Symptoms of autonomic dysfunction can be helpful in directing attention toward specific neuropathies that have prominent autonomic symptoms. It is important to ask about orthostatic intolerance (lightheadedness, presyncopal symptoms or syncope), reduced or excessive sweating, heat intolerance, and bladder, bowel, and sexual dysfunctions. Anorexia, early satiety, nausea, and vomiting are symptoms suggestive of gastroparesis. The degree of autonomic involvement can be documented by noninvasive autonomic function studies (see Chapter 77).

Historical information regarding onset, duration, and evolution of symptoms provides important clues to diagnosis. Knowledge about the time course of disease (acute, subacute, or chronic) and the course (monophasic, progressive, or relapsing) narrows diagnostic possibilities. Guillain-Barré syndrome (GBS), acute porphyria, vasculitis, neuralgic amyotrophy and some forms of toxic neuropathies have acute presentations. A relapsing course is found in chronic inflammatory demyelinating polyradiculoneuropathy (CIDP), acute porphyria, Refsum disease, hereditary neuropathy with liability to pressure palsies (HNPP), familial brachial plexus neuropathy, and repeated episodes of toxin exposure.

In patients with a chronic indolent course over many years, inquiries about similar symptoms and bony deformities (such as pes cavus) in immediate relatives often point to a familial polyneuropathy. Inherited polyneuropathies are a major cause of undiagnosed polyneuropathies, accounting for about 30% of patients referred to tertiary centers for diagnosis. Molecular genetic testing or the clinical and electrophysiological evaluation of relatives of patients with undiagnosed neuropathy may corroborate that the disorder is familial. The presence of constitutional symptoms such as weight loss, malaise, and anorexia suggests an underlying systemic disorder as a cause of the polyneuropathy. Inquiry should be made about preceding or concurrent associated medical conditions (diabetes mellitus, hypothyroidism, chronic renal failure, liver disease, intestinal malabsorption, malignancy, connective tissue diseases, human immunodeficiency virus [HIV] seropositivity); drug use, including over-the-counter vitamin preparations (vitamin B₆); alcohol and dietary habits; and exposure to solvents, pesticides, or heavy metals.

Diagnostic Clues from the Examination

The first step in the examination of patients with neuropathy is to determine the anatomical pattern and localization of the disease process and whether motor, sensory, or autonomic nerves are involved.

In mononeuropathy, the neurological deficit follows the distribution of a single nerve. For example, in a patient with foot drop due to a common fibular (peroneal) nerve lesion, the neurological examination reveals weakness of ankle and toe dorsiflexion and ankle eversion, but ankle inversion, toe flexion, and plantar flexion are normal, since muscles controlling these functions are innervated by the tibial nerve. Similarly, sensory loss is restricted to the lower two-thirds of the lateral leg and dorsum of the foot, but sensation on the sole of the foot is normal.

In multiple mononeuropathies (mononeuropathy multiplex), the neurological findings should point to simultaneous or sequential damage to two or more noncontiguous peripheral nerves. Confluent multiple mononeuropathies, such as with involvement of the fibular and tibial nerves or median and ulnar nerves, may give rise to motor weakness with sensory loss that can simulate a length-dependent peripheral polyneuropathy. EDX studies ascertain whether the primary pathological process is axonal degeneration or segmental demyelination (Box 76.2). Approximately two-thirds of patients with multiple mononeuropathies display a picture of axonal damage. Ischemia caused by systemic or nonsystemic vasculitis or microangiopathy in diabetes mellitus should be considered. Other less common causes are disorders affecting interstitial structures of nerve, namely infectious, granulomatous, leukemic, or neoplastic infiltration, including Hansen disease (leprosy) and sarcoidosis. In the event focal demyelination or motor conduction block leads to multiple mononeuropathies, multifocal acquired demyelinating sensory and motor neuropathy (Lewis-Sumner syndrome), multifocal motor neuropathy, or HNPP should be considered.

In polyneuropathy, the sensory deficits generally follow a length-dependent stocking-glove pattern. By the time sensory disturbances of the longest nerves in the body (lower limbs) have reached the level of the knees, paresthesias are usually noted in the distribution of the second-longest nerves (i.e., those in the upper limbs) at the tips of the fingers. When sensory impairment reaches the midthigh, involvement of the third-longest nerves, the anterior intercostal and lumbar segmental nerves, gives rise to a tent-shaped area of hypoesthesia on the anterior chest and abdomen. Involvement of the recurrent laryngeal nerves may occur at this stage, with hoarseness. Motor weakness follows a dying-back pattern and usually is greater in extensor foot muscles than in corresponding flexors. For example, heel walking is affected earlier than toe walking in most polyneuropathies. It is helpful to determine the relative extent of sensory, motor, and autonomic fiber involvement, although most polyneuropathies produce mixed sensorimotor deficits and some degree of autonomic dysfunction.

Motor deficits tend to dominate the clinical picture in acute and chronic inflammatory demyelinating polyneuropathies, hereditary motor and sensory neuropathies, and in neuropathies associated with osteosclerotic myeloma, porphyria, lead toxicity, organophosphate intoxication, and hypoglycemia (Box 76.3). The distribution of weakness provides important information. Asymmetrical weakness without sensory loss suggests a motor neuronopathy such as motor neuron disease or multifocal motor neuropathy. The facial nerve can be affected in several peripheral nerve disorders (Box 76.4). In most polyneuropathies, the legs are more severely affected than the arms, with several notable exceptions (Box 76.5). Polyradiculoneuropathies cause both proximal and distal muscle weakness. For example, proximal and distal weakness is encountered in acute and chronic inflammatory demyelinating polyradiculoneuropathies, osteosclerotic myeloma, porphyria, and diabetic lumbar radiculoplexopathy. Nerve root involvement is confirmed by denervation in paraspinal muscles on needle EMG.

Autonomic dysfunction of clinical importance is seen in association with specific acute (e.g., GBS) or chronic (e.g., amyloid and diabetic) sensorimotor polyneuropathies. Rarely, an autonomic neuropathy can be the exclusive manifestation of a peripheral nerve disorder, without somatic nerve involvement (Box 76.6).

Predominant sensory involvement may be a feature of polyneuropathies caused by diabetes, carcinoma, Sjögren syndrome, dysproteinemia, acquired immunodeficiency syndrome (AIDS), vitamin B₁₂ deficiency, celiac disease, inherited and idiopathic sensory neuropathies, and intoxications with cisplatin, thalidomide, or pyridoxine. Loss of sensation in peripheral neuropathies often involves all sensory modalities. However, the impairment may be restricted to selective sensory modalities in many situations, which makes it possible to correlate the type of sensory loss with the diameter size of affected afferent fibers (Fig. 76.4). Pain and temperature sensation are mediated by unmyelinated and small myelinated Aδ fibers, whereas vibratory sense, proprioception, and the afferent limb of the tendon reflex are subserved by large myelinated Aα and Aβ fibers. Light touch is mediated by both large and small myelinated fibers. In polyneuropathies preferentially affecting small fibers, diminished pain and temperature sensation predominate, along with spontaneous burning pain, painful dysesthesias, and autonomic dysfunction. There is preservation of tendon reflexes, balance, and motor strength, and hence few abnormal objective neurological signs are found on examination. A pattern of sensory loss that is very characteristic is distal loss of pinprick sensation, above which is a band of hyperalgesia (exaggerated pain from noxious stimuli), with normal sensation above this level. Relatively few disorders cause selective small-fiber neuropathies (Mendell and Sahenk, 2003) (Box 76.7). Selective large-fiber sensory loss is characterized by areflexia, sensory ataxia, and loss of joint position and vibration sense. Loss of joint position may also manifest as pseudoathetosis (involuntary sinuous movements of fingers and hands when the arms are outstretched and the eyes are closed) and/or a Romberg sign (disproportionate loss of balance with eyes closed compared with eyes open). Striking sensory ataxia, together with pseudoathetosis or asymmetrical truncal or facial sensory loss, directs attention to a primary disorder of sensory neurons or polyganglionopathies. The differential diagnosis of ataxic sensory neuropathies is limited (Box 76.8).

Fig. 76.4 Myelinated fiber (MF) and unmyelinated fiber (UF) size-frequency histograms of a normal sural nerve. Fiber size distribution is bimodal for MF but unimodal for UF. MF density in normal sural nerve ranges from 6000 to 10,000 fibers/mm² of fascicular area. Number of UFs is normally about four times that of MFs. Corresponding compound nerve action potential recorded from the sural nerve in vitro is shown at top. Three distinct peaks indicated by arrows are, from left to right, Aa, Aa, and C-potentials, which correspond to large MF, small MF, and UF peaks, respectively.

Palpation of peripheral nerves is an important though unreliable part of the examination. Hypertrophy of a single nerve trunk suggests either a neoplastic process (e.g., neurofibroma, schwannoma, malignant nerve sheath tumor) or localized perineurial hypertrophic neuropathy. Generalized or multifocal nerve hypertrophy is found in a limited number of peripheral nerve disorders including leprosy, neurofibromatosis, Charcot-Marie-Tooth (CMT) disease types 1 and 3, acromegaly, Refsum disease, and rarely CIDP.

Certain telltale signs of the skin and its appendages may direct the experienced examiner to a specific diagnosis (Table 76.1): alopecia is seen in thallium poisoning, tightly curled hair in giant axonal neuropathy; white transverse nail bands termed Mees lines in arsenic or thallium intoxications; purpuric skin eruptions of the legs in cryoglobulinemia and some vasculitides; skin hyperpigmentation or hypertrichosis in POEMS syndrome (polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy, and skin changes); telangiectasias over the abdomen and buttocks in Fabry disease; enlarged yellow-orange tonsils in Tangier disease; pes cavus and hammer toes in CMT disease; and overriding toes and ichthyosis in Refsum disease.

Table 76.1 Neuropathies with Skin, Nail, or Hair Manifestations

POEMS, Polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy, and skin changes.

Electrodiagnostic Studies

It is helpful to follow a decision-making pathway based initially on the overall pattern of distribution of deficits, followed by the electrophysiological findings, and finally the clinical course (Fig. 76.5). EDX studies, carefully performed and directed to the particular clinical situation, play a key role in the evaluation by (1) confirming the presence of neuropathy, (2) precisely locating focal nerve lesions, and (3) giving information as to the nature of the underlying nerve pathology (Gooch and Weimer, 2007; Wilbourn, A.J., 2002) (see Chapter 32B).

Fig. 76.5 Diagnostic approach to evaluation of a patient with peripheral neuropathy. Electromyography (EMG) denotes electrodiagnostic studies. DNA diagnostic testing or specific biochemical tests are available for conditions marked with asterisks. CIDP, Chronic inflammatory demyelinating polyradiculoneuropathy; CMT, Charcot-Marie-Tooth disease; CMTX, Charcot-Marie-Tooth disease X-linked; GBS, Guillain-Barré syndrome; HNPP, hereditary liability to pressure palsies; NP, neuropathy.

Because routine sensory nerve conduction studies assess only large myelinated fibers, such studies may be entirely normal in selective small fiber neuropathies. Quantitative sensory testing assessing cold and heat-pain thresholds, tests of sudomotor function, and skin biopsy with analysis of intraepidermal nerve fiber density may be helpful in confirming the unmyelinated nerve fiber abnormalities (Devigili et al., 2008). Since sweating mediated by unmyelinated sympathetic cholinergic fibers is often impaired, the quantitative sudomotor axon reflex (QSART) that evaluates sweating is a highly specific and sensitive method (sensitivity of 80%) to confirm small nerve fiber damage. Quantitative sensory testing assessing both vibratory and thermal detection thresholds has become a useful addition to the bedside sensory examination in controlled clinical trials. Its use in routine clinical practice remains limited because the test is still subjective in that it requires patient cooperation and is time consuming.

Nerve and Skin Biopsy

Skin punch or blister biopsies that demonstrate loss of intraepidermal nerve fibers is an alternative method for documenting small fiber neuropathy (Panoutsopoulou, 2009). Only unmyelinated intraepidermal networks of nerve fibers can be demonstrated by immunostaining with the panaxonal marker protein gene product 9.5, studied best with the use of confocal microscopy. Age, gender, and site of skin biopsy have a profound effect on epidermal nerve fiber density. The density of intraepidermal nerve fibers is reduced in skin biopsies obtained from patients with idiopathic, HIV-associated, diabetic, and other sensory neuropathies (Kennedy, 2004). Skin punch biopsy is most useful in patients with suspected small fiber neuropathy, when nerve conduction studies, are normal. The diagnosis of small-fiber neuropathy is best accomplished when at least two abnormal results are met, including positive clinical findings, quantitative sensory testing, QSART, and skin biopsy examinations (Devigili et al., 2008). Skin punch biopsy only detects the presence of skin nerve abnormalities and rarely leads to a specific etiological diagnosis. The skin biopsy also does not permit the study of myelinated fibers unless a thicker biopsy including dermis is obtained. Finally, unlike sural nerve biopsy, the interstitial pathological processes of the nerves cannot be studied.

Nerve biopsy (other than for the diagnosis of vasculitis and neoplasia) should be performed only in centers with established experience with the surgical procedure, handling of nerve specimens, and pathological technique; otherwise little useful information is likely to be obtained. The sural nerve is selected most commonly for biopsy, because the resultant sensory deficit is restricted to a small area over the heel and dorsolateral aspect of the foot, and because its morphology has been well characterized in health and disease. The superficial fibular nerve is an alternative lower-extremity cutaneous nerve suitable for biopsy and has the advantage of allowing simultaneous biopsy of the peroneus brevis muscle through the same incision. This combined distal nerve and muscle biopsy procedure increases the yield of identifying suspected vasculitis (Collins et al., 2000; Vital et al., 2006). In contrast, adding a proximal muscle (e.g., quadriceps) to a cutaneous nerve biopsy (e.g., sural) does not significantly increase the diagnostic yield compared to nerve biopsy alone (Bennett et al., 2008). In patients with proximal involvement of the lower limbs, the intermediate cutaneous nerve of the thigh combined with a muscle biopsy can be performed. When the risk of complication is increased from a biopsy of the lower limbs (e.g., in significant distal leg ischemia, edema) or the neuropathy is preferentially more pronounced in the upper limbs, a cutaneous nerve biopsy of superficial radial or antebrachial nerves may be performed. When the imaging studies indicate a plexus or nerve root pathological process (e.g., inflammatory, infiltrative), a fascicular biopsy of the affected nerve by an expert surgeon may provide invaluable information. Nerve biopsy has proved to be particularly informative when techniques such as single teased fiber preparations, semithin sections, ultrastructural studies, and morphometry are applied to quantitate the nerve fiber pathology. Nowadays, relatively few disorders remain in which a nerve biopsy is essential for diagnosis (Pleasure, 2007; Said, 2002) (Box 76.9). In general, nerve biopsy is most frequently diagnostic in suspected vasculitis, amyloid neuropathy, and leprosy. It is helpful in the recognition of CIDP, inherited disorders of myelin, and some rare axonopathies in which distinctive axonal changes occur, such as in giant axonal neuropathy and polyglucosan body disease. The availability of molecular genetic tests for several CMT neuropathies, HNPP, and familial transthyretin amyloidosis has decreased the necessity for nerve biopsy in these conditions.

Nerve biopsy is an invasive procedure and is associated with as high as 15% complication rate—particularly minor wound infections, wound dehiscence, and stump neuromas. Approximately one-third of patients (particularly those without much sensory loss initially) report unpleasant sensory symptoms at the sural nerve biopsy site that are still present 1 year after the biopsy (Gabriel et al., 2000). The area of the original sensory deficit declines by 90% after 18 months because of collateral reinnervation (Theriault et al., 1998). The complications may be greater if substantial foot ischemia is present or if the patient smokes cigarettes.

Other Laboratory Tests

The clinical neuropathic patterns and the results of EDX studies guide the experienced clinician to select the most appropriate laboratory tests. A few laboratory tests should be obtained routinely in all patients with peripheral polyneuropathy. These include complete blood cell count (CBC), sedimentation rate (with or without C-reactive protein), renal functions, fasting blood sugar, thyroid studies, vitamin B₁₂ level, and serum protein electrophoresis with immunofixation electrophoresis. It is important to screen for monoclonal proteins in all patients with chronic undiagnosed polyneuropathy, particularly those older than 60 years, because 10% of such patients have a monoclonal gammopathy. Cerebrospinal fluid (CSF) examination is helpful in the evaluation of suspected demyelinating neuropathies and polyradiculopathies related to meningeal carcinomatosis or lymphomatosis.

Several serum autoantibodies with reactivity to various components of peripheral nerve have been associated with peripheral neuropathy syndromes, and reference laboratories offer panels of nerve antibodies for sensory, sensorimotor, and motor neuropathies. It must be emphasized that the clinical relevance of most autoantibodies has not been established for patient treatment, and their use is not cost-effective (Vernino and Wolfe, 2007). Those of greatest clinical utility are listed in Table 76.2 (Kissel, 1998). An ever-increasing number of molecular genetic tests for inherited neuropathies is available at reference laboratories (see Hereditary Neuropathies, later).

Table 76.2 Neuropathies Associated with Serum Autoantibodies

Ig, Immunoglobulin.

In patients with initially undiagnosed peripheral neuropathy referred to specialized centers, a definite diagnosis can be made in more than 75% of cases. Inherited neuropathies, CIDP, and neuropathies associated with other systemic diseases accounted for most diagnoses. The improved diagnostic rate resulted in large measure from detailed clinical, EDX and laboratory evaluations and study of relatives of patients with undiagnosed neuropathy.

Definition and Classification of Mononeuropathies

Mononeuropathy is defined as a disorder of a single peripheral nerve. This may result from compression, traction, laceration, thermal, or chemical injury. The damage may involve one or more structural components of the peripheral nerve, while the pathophysiological responses to peripheral nerve lesions include axon loss, demyelination, or a combination of both.

Peripheral nerve injuries are classified based on functional status of the nerve and histological findings. Seddon divided peripheral nerve injury into three classes: neurapraxia, axonotmesis, and neurotmesis. This classification remains popular, particularly among surgeons, because of its correlation to outcome (Seddon, 1975). Later, Sunderland (1991) revised the classification into five degrees that have better prognostic implications.

Mononeuropathies of the Upper Extremities

Entrapment neuropathies of the upper extremities are shown in Table 76.3.

Median Nerve

Applied Anatomy

The median nerve, formed from contributions of the lateral cord (C6 and C7 fibers) and medial cord (C8 and T1 fibers) of the brachial plexus, runs into the forearm between the two heads of the pronator teres. It then gives off branches to the pronator teres, flexor carpi radialis, flexor digitorum sublimis, and the palmaris longus muscles and the anterior interosseous nerve. The anterior interosseous nerve is the largest branch of the median nerve and is a pure motor nerve. It arises from the median nerve distal to these motor branches in the upper forearm and innervates the flexor pollicis longus, pronator quadratus, and flexor digitorum profundus muscles of the index and middle fingers. The median nerve then enters the wrist through the carpal tunnel, formed by the carpal bones and the transverse carpal ligament, the latter forming its roof. Before reaching the wrist, the median nerve gives off the palmar cutaneous sensory branch which runs subcutaneously (not through the carpal tunnel) to innervate the skin over the thenar eminence. Distal to the carpal tunnel, the median nerve divides into motor and sensory divisions. The motor division innervates the first and second lumbricals and most muscles of the thenar eminence including the opponens pollicis, abductor pollicis brevis, and superficial head of the flexor pollicis brevis. The sensory fibers of the median nerve innervate the skin of the thumb, index, middle, and lateral half of the ring fingers.

Median Nerve Entrapment at the Wrist (Carpal Tunnel Syndrome)

Carpal tunnel syndrome (CTS) is by far the most common entrapment neuropathy. This entrapment occurs in the tunnel through which the median nerve and long finger flexor tendons pass. Because the transverse carpal ligament is an unyielding fibrous structure forming the roof of the tunnel, tenosynovitis or arthritis in this area often produces pressure on the median nerve. The syndrome is frequently bilateral and usually of greater intensity in the dominant hand.

Symptoms consist of nocturnal pain and paresthesias, most often confined to the thumb, index, and middle fingers, but may be reported to involve the entire hand. Patients complain of tingling numbness and burning sensations, often awakening them from sleep. Referred pain may radiate to the forearm and even as high as the shoulder (Stevens et al., 1999). Symptoms are often provoked after excessive use of the hand or wrist or during ordinary activities such as driving or holding a phone, book, or newspaper, in which the wrist is assumed in either a flexed or extended posture. Objective sensory changes may be found in the distribution of the median nerve, most often impaired two-point discrimination, pinprick and light touch sensation, or occasionally hyperesthesia in the thumb, index, and middle fingers, with sparing of the skin over the thenar eminence. Thenar (abductor pollicis brevis muscle) weakness and atrophy may be present in advanced cases of CTS (Fig. 76.6). Flexing the patient’s hand at the wrist for 1 minute (Phalen maneuver) or hyperextension of the wrist (reversed Phalen maneuver) can reproduce the symptoms, is present in about 80% of patients, and is rarely false positive. A positive Tinel sign, in which percussion of the nerve at the carpal tunnel causes paresthesias in the distribution of the distal distribution of the median nerve, is present in approximately 60% of affected patients but is not specific for CTS and may be false positive.

Fig. 76.6 Thenar atrophy in chronic bilateral carpal tunnel syndrome.

Work-related wrist and hand symptoms (repetitive motion injury) from cumulative trauma in the workplace have received increasing attention by the general public in recent years (Thomsen et al., 2002). Although a proportion of these cases have bona fide CTS, longitudinal natural history data suggest that the majority of industrial workers do not develop symptoms of CTS (Nathan et al., 1998). Symptoms consistent with hand and wrist arthritis in a variety of occupational settings are now recognized as being much more common than CTS (Dillon et al., 2002). CTS appears to occur in work settings that include repetitive forceful grasping or pinching, awkward positions of the hand and wrist, direct pressure over the carpal tunnel, and the use of handheld vibrating tools. Increased risk for the syndrome has been found in meat packers, garment workers, butchers, grocery checkers, electronic assembly workers, musicians, dental hygienists, and housekeepers. The highest reported incidence of work-related CTS, based on the number of carpal tunnel surgeries performed, was 15% among a group of meat packers. Although computer keyboard use has long been thought to be related to developing carpal tunnel symptoms, recent data provide no convincing correlation between intensive keyboard use and the subsequent development of CTS (Papanicolaou et al., 2001; Stevens et al., 2001).

Diseases and conditions that have been found to predispose to the development of CTS include pregnancy, diabetes, obesity, age, rheumatoid arthritis, hypothyroidism, amyloidosis, gout, acromegaly, certain mucopolysaccharidoses, arteriovenous shunts for hemodialysis, old fractures at the wrist, and inflammatory diseases involving tendons or connective tissues at the wrist level (Becker et al., 2002). On rare occasions, CTS may be familial, and some patients with CTS have carpal canals that are significantly narrower than average.

The most commonly performed EDX test for CTS are the median nerve sensory and motor conduction studies, which exhibit delayed sensory or motor latencies across the wrist in about 70% of patients. However, these studies are not sensitive enough in the diagnosis of CTS and fail to detect up to a third of patients with CTS, particularly those with mild and early symptoms. Recording the median latency at short distances over the course of the median nerve from palm to wrist and/or comparing this latency with the latency for the ulnar or radial nerve at the same distance (internal comparison nerve conduction studies) increase the sensitivity of these nerve conduction studies (Stevens, 1997) (Table 76.5).

Table 76.5 Internal Comparison Nerve Conduction Studies in the Evaluation of Carpal Tunnel Syndrome

Mild CTS must be distinguished from proximal median neuropathies, upper brachial plexopathy, C6 or C7 radiculopathies, and polyneuropathy involving the hands. Occasionally, a transient ischemic attack may mimic the symptoms of CTS.

In cases with only mild sensory symptoms, treatment with splints in neutral position, nonsteroidal antiinflammatory drugs (NSAIDs), and local corticosteroid injection often suffice. Withdrawal of provoking factors is also important. Although nonoperative treatments have been advocated (Osterman et al., 2002), a comparison of splinting versus surgery suggested that the latter may have a better long-term outcome than the former (Gerritsen et al., 2002). Use of a range of devices and appliances to protect the hand against CTS, including gel-padded gloves, has shown little if any improvement in objective measures of nerve function. Severe sensory loss and thenar atrophy suggest the need for surgical carpal tunnel release. Open surgical sectioning of the volar carpal ligament or fiberoptic techniques are often successful, with more than 90% of patients having prompt resolution of pain and paresthesias (Mirza and King, 1996). Improvement in distal latencies may lag behind the relief of symptoms. Comparing with preoperative values, nerve conduction studies demonstrate improvement in those with moderate abnormalities preoperatively, whereas patients with severe or no abnormalities on baseline nerve conduction studies had poorer results (Bland, 2001). A correlation between patients seeking workers’ compensation who hire attorneys and poorer operative outcomes has also been reported (Katz et al., 2001a). Older individuals may not improve as much as younger patients (Porter et al., 2002), and factors such as poor mental health, significant alcohol consumption, longer disease duration, and male gender also portend a poorer outcome. Rarely, symptoms persist after operation. Poor surgical results usually are associated with incomplete sectioning of the transverse ligament, surgical damage of the palmar cutaneous branch of the median nerve by an improperly placed skin incision, scarring within the carpal tunnel, or an incorrect preoperative diagnosis. Surgical reexploration may be required in diagnostically certain cases with poor response to the initial operation (Steyers, 2002).

Median Nerve Compressions at the Elbow

Anterior Interosseous Nerve Syndrome

In the relatively rare anterior interosseous nerve syndrome, the nerve may be compressed by fibrous bands attached to the flexor digitorum superficialis muscle, an anomalous muscle such as accessory head of the flexor pollicis longus (Gantzer muscle), or by an external compression such as with anterior elbow dislocations or complex elbow fractures. Isolated involvement of this nerve more often may suggest a partial neuralgic amyotrophy (idiopathic brachial plexus neuropathy, Parsonage-Turner syndrome) or less commonly a restricted form of multifocal motor neuropathy with conduction block. However, a careful and detailed electrophysiological study may reveal involvement of other nerves. Patients often complain of an acute or subacute onset of pain in the forearm or elbow. The patient is unable to flex the distal phalanges of the thumb and index finger, making it impossible to form a circle with those fingers (pinch or O sign). Sensory and motor nerve conduction studies of the median nerve are usually normal. Needle EMG reveals denervation in muscles innervated by the anterior interosseous nerve, including the flexor pollicis longus, pronator quadratus, and flexor digitorum profundus muscles of the index and middle fingers. Spontaneous recovery usually occurs within 3 to 12 months, and therefore surgery may not be necessary unless penetrating injury, fracture, or progressive deterioration and weakness are detected.

Pronator Teres Syndrome

In the pronator teres syndrome, the median nerve is compressed in the proximal forearm between the two heads of pronator teres muscle, a fibrous arcade of the flexor digitorum superficialis muscle, or the lacertus fibrosus (a thick fascial band extending from the biceps tendon to the forearm fascia). This extremely rare and controversial entrapment may develop in individuals engaged in repetitive pronating movements of the forearm. Patients usually experience a vague aching pain in the volar aspect of the elbow and forearm, beginning or worsening during activities involving grasping or pronation or both. There is also an insidious onset of paresthesias and numbness of the palm of hand, mimicking CTS but without the nocturnal symptoms. Resistance to pronation produces pain in the proximal forearm. The pronator teres may be firm and tender on palpation, and the Tinel sign may be elicited over the median nerve in the region of the elbow. Weakness of median-innervated muscles such as the flexor pollicis longus, pronator quadratus, and abductor pollicis brevis (but not of pronator teres) is rarely demonstrated, in contrast to traumatic cases such as following elbow dislocation, forearm fracture, or intracompartmental hemorrhage. Nerve conduction studies in the median nerve are usually normal and do not show the distal median motor and sensory latencies at the wrist that accompany CTS. Needle EMG is also usually normal, with no definite signs of denervation. Injection of corticosteroids into the pronator teres muscle, NSAIDs, and immobilization of the arm with the elbow flexed to 90 degrees and in mild pronation often provide relief of symptoms. On occasion, surgery is controversial but may be necessary, but patients may gain only partial relief.

Median Nerve Entrapment at the Ligament of Struthers

An often bilateral supracondylar spur of the humerus is present in approximately 1% of normal individuals. This beadlike bony or cartilaginous projection arises from the anteromedial surface of the humerus, located about 5 cm above the medial epicondyle. A fibrous band, the ligament of Struthers, extends from this spur to the medial epicondyle and may rarely compromise the median nerve and the brachial artery above the elbow. Clinical symptoms resemble the pronator teres syndrome, but the radial pulse diminishes when the forearm is fully extended in supination because of the concomitant entrapment of the brachial artery. Elbow extension causes aggravation of the pain. On needle EMG, there is usually denervation in the abductor pollicis brevis, flexor pollicis longus, pronator quadratus, and pronator teres. Involvement of the pronator teres muscle theoretically allows differentiation of the ligament of Struthers syndrome from the pronator teres syndrome. Treatment consists of surgical excision of the spur and ligament.