Skin biopsy for the evaluation of peripheral nerve disease

INTRODUCTION — The study of peripheral nerve disorders has been aided by the pathologic analysis of tissue biopsy of several anatomic structures, including muscle, nerve, and skin. Tissue biopsy can improve diagnostic yield in selected cases and can be reasonably and safely undertaken without major morbidity.

This chapter will review the technique of skin biopsy in the evaluation and study of peripheral nerve disease. Muscle and nerve biopsy are discussed separately. (See "Approach to the patient with muscle weakness" and "Overview of polyneuropathy", section on 'Nerve biopsy'.)

Other diagnostic methods for the evaluation of neuropathy and neuromuscular disease are discussed elsewhere. (See "Overview of nerve conduction studies" and "Overview of electromyography".)

TECHNIQUE — The existence of epidermal innervation was controversial for much of the 20^th century, despite the fact that Langerhans first recognized epidermal nerve fibers in the 1860s [1]. His drawings provide an accurate representation of these epidermal fibers, as confirmed by modern immunohistochemical techniques [2]. Despite their early recognition, epidermal nerves were difficult to visualize using standardized histochemical techniques, so little effort was made to study them for most of the 20^th century [3]. Elegant studies by Winkelman and Dyck demonstrated abnormalities of cutaneous receptors, such as Meissner corpuscles, in subjects with inherited neuropathies, but this technique was of little diagnostic utility [4].

In 1989, an immunohistochemical technique was developed that permitted relatively easy visualization of cutaneous nerves and sensory organs [2,5,6]. This technique has facilitated widespread study of cutaneous innervation for both clinical and research applications.

Normal anatomy — A basic appreciation of normal dermal anatomy is essential for understanding pathologic abnormalities [7].

●Skin categories and constituents – Skin areas are classified as being glabrous (hairless) or nonglabrous:

•Glabrous skin is located on the palms, soles, and fingertips and contains more sweat glands and cutaneous sensory organs dedicated to touch, position, and vibration sensation. Pacinian and Meissner corpuscles, Merkel cells, and abundant sweat glands are easily observed. Small myelinated fibers that supply these structures are also seen within the dermis.

•Nonglabrous skin has fewer sensory organs and myelinated dermal fibers than glabrous skin and a greater number of hair follicles and piloerector muscles, which are supplied by small autonomic nerves.

Skin in both areas has abundant epidermal nerves in normal subjects (picture 1).

●Cutaneous nerve fibers – The dermis extends into the epidermis in undulations called papillae. Normally, there is a network of bundled dermal nerve fibers beneath the dermal-epidermal junction. Some sensory organs such as Merkel cells and Meissner corpuscles are located at the apex of dermal papillae (picture 2). Nerve fibers supplying the epidermis also enter in this region, where they lose their myelin sheath.

Small thinly myelinated (A-delta) and unmyelinated (C) fibers are responsible for pain and temperature sensation. A single fiber typically divides into several branches immediately after crossing into the epidermis. These branches extend in parallel fashion to near the surface of the skin. Independent of age, the density of epidermal innervation is greater at more proximal sites on the body (eg, the back and thigh) than it is at distal locations (eg, the ankle) [8,9].

Methodology — Skin biopsy is a benign procedure. Most laboratories use a 3 mm punch to obtain skin tissue. This does not require a suture, and the biopsy site typically heals on its own without additional intervention over several weeks. Risks include local bleeding and infection. Both are uncommon and typically minor. The procedure is well tolerated and may be repeated multiple times in order to follow neuropathy progression.

●Biopsy site – The biopsy site depends on the specific indication [10]. For length-dependent peripheral neuropathies, most laboratories recommend biopsy at the distal leg (8 cm proximal to the lateral malleolus) and at a more proximal site such as the lateral thigh (10 cm distal to the iliac crest).

●Tissue preparation – The tissue is immediately placed in either 2% periodate-lysine-paraformaldehyde (PLP) or Zamboni solution (2% paraformaldehyde, picric acid) for 12 to 24 hours at 4°C and is then placed in a cryoprotectant solution. The biopsies can then be stored in a freezer or shipped to a central laboratory for processing. The punch biopsy is cut into thick (eg, 50 micron) sections and stained with antiprotein gene product 9.5 antibody (PGP 9.5), which binds to ubiquitin hydrolase [5]. PGP 9.5 avidly stains all axons [2].

Pathologic analysis — In normal subjects, both epidermal and dermal structures are readily observed and evaluated. The number and morphology of axons within the epidermis are evaluated, by either qualitative or quantitative assessment, to determine intraepidermal nerve fiber density (IENFD) [11]. These values are compared with age-dependent normal values [8,12].

Patients with symptoms isolated to the feet or toes typically have severe depletion or absence of intraepidermal nerve fibers at the level of the ankle or distal leg on skin biopsy (picture 3). Reduced IENFD is often appreciated at the level of the proximal thigh. In addition to reduced fiber density, morphologic abnormalities are often observed. Axonal swellings, defined as focal enlargements greater than two times the width of the axon, and complex branching patterns with fibers traversing parallel to the skin surface may be a pre-degenerative finding (picture 4) [13,14]. Care must be taken in interpreting isolated morphologic abnormalities, as small axonal swellings and beading are often observed in normal subjects and may be influenced by technical factors. However, the presence of large axonal swellings has been associated with progressive decline in fiber density among subjects with suspected small fiber neuropathy, and their presence in the proper clinical setting is supportive of that diagnosis.

There are two different methods of microscopic analysis. Many groups use standard bright field light microscopy. Its primary advantages are convenience, cost, and ease of use. Other centers use confocal microscopy. Confocal imaging is more labor intensive and expensive, but it allows for multiple antibodies to be more readily used. For example, an antibody to collagen IV may be used to stain the basement membrane with a different fluorescent color, so points where the nerves cross are more easily appreciated.

For clinical interpretation, both microscopic techniques are adequate. For some research applications, confocal imaging is advantageous. It is important that the ordering clinician knows which technique is being used, particularly if formal quantification of IENFD is being performed, because densities are somewhat higher using confocal imaging [15].

CLINICAL APPLICATIONS

Small fiber neuropathy — The most frequent clinical indication for skin biopsy in the evaluation of peripheral nerve disease is for the diagnosis of patients with suspected small fiber sensory neuropathy. In disorders affecting mainly small, unmyelinated nerve fibers, standard electrophysiologic testing is often normal, and sural nerve biopsy may be normal or only minimally abnormal.

●Clinical features – Patients with small fiber sensory neuropathy commonly have severe complaints but only minimal objective abnormalities on examination. Typical symptoms of small fiber neuropathy are distal burning, pain, numbness, and paresthesia and, frequently, autonomic symptoms [16-20]. Most patients have a combination of both negative signs (eg, pinprick and thermal hypoesthesia) and positive signs (eg, allodynia, hyperalgesia) [21].

●Diagnostic guidelines – American guidelines regarding the evaluation of distal symmetric polyneuropathy endorse intraepidermal nerve fiber density (IENFD) determination by skin biopsy, using antiprotein gene product 9.5 immunohistochemistry, as a validated and reproducible marker of small fiber sensory neuropathy [22]. Similarly, European guidelines conclude that distal leg skin biopsy with quantification of IENFD is a reliable and efficient technique to assess the diagnosis of small fiber neuropathy [23]. In one study of 67 patients with a pure small fiber neuropathy, skin biopsy was diagnostic in 88 percent of cases, compared with a diagnostic rate by clinical examination of 55 percent [24]. Proposed diagnostic criteria for small fiber neuropathy require a combination of clinical signs (eg, pinprick and thermal sensory loss, allodynia, hyperalgesia) and abnormal IENFD or quantitative sensory testing [21].

●Ancillary testing techniques – Alternative tests for small fiber function are limited in various ways. Corneal confocal microscopy is a rapid and noninvasive method for directly measuring nerve fiber density in the cornea but it is not widely available and may have a lower diagnostic sensitivity than other small fiber tests [25,26]. Sympathetic skin response is very insensitive for the diagnosis of small fiber neuropathy [27,28]. Quantitative sensory testing for heat, pain, or cold detection thresholds is sensitive, but subjects with psychogenic sensory complaints may have convincing abnormalities [29]. Electrochemical skin conductance is incompletely validated and not widely available [30]. Tests of sudomotor function, particularly the quantitative sudomotor axonal reflex test, are sensitive and specific [22] but have lower reproducibility compared with quantitative sensory testing [31] and require special equipment and expertise [32]. By contrast, skin biopsy has a very high diagnostic accuracy, with positive and negative predictive values of greater than 90 percent [33].

Skin biopsy typically does not allow for differentiation of different causes of small fiber neuropathy. Reduced fiber density has been documented in a variety of neuropathies, including metabolic, infectious, inflammatory, environmental, hereditary, nutritional, and idiopathic neuropathies [11,34-42]. If amyloidosis is suspected, biopsy sections may be stained with Congo red to detect amyloid deposits [43].

In patients with foot pain and possible neuropathy, a skin biopsy with normal fiber density and morphology should prompt a careful search for a central nervous system or orthopedic problem. In a prospective series of 117 subjects with foot pain and suspected neuropathy, normal nerve conduction studies and skin biopsy results were noted in 13 patients (11 percent) [44]. In this group, two were ultimately diagnosed with multiple sclerosis.

Painful sensory ganglionopathies — Skin biopsy may be particularly useful in subjects who have pain and numbness with diffuse symptoms that do not conform to a length-dependent pattern. While this situation most often suggests non-neuropathic disease or a psychogenic disorder, rare patients can have a non-length-dependent small fiber ganglionopathy rather than the classic distal symmetric/length-dependent form of small fiber neuropathy [45]. Such patients can present with diffuse, often acute, burning pain [46]. Multiple conditions have been associated with non-length-dependent painful sensory neuropathies, including but not limited to diabetes, impaired glucose tolerance, Sjögren's disease, hepatitis C, and celiac disease [47]. In a study of patients with small fiber ganglionopathy of various etiologies who had painful non-length-dependent sensory symptoms, skin biopsy was abnormal in 14 of 17 cases [45].

Distal symmetric polyneuropathy — Skin biopsy can be used for the diagnosis of distal symmetric polyneuropathy, including those due to the involvement of small unmyelinated or large myelinated fibers, when the diagnosis is uncertain after initial testing. In most cases, skin biopsy is seldom needed for the diagnosis of large, myelinated fiber neuropathy because less invasive electrodiagnostic studies (ie, nerve conduction studies and electromyography) are generally obtained first and are sensitive for the detection of large fiber neuropathy. By contrast, electrodiagnostic studies are insensitive to small and unmyelinated fiber injury.

Other conditions — Skin biopsy may be used in other conditions where initial testing is nondiagnostic. Such conditions include:

●Mixed fiber neuropathies due to idiopathic or diabetic neuropathies may be diagnosed early with skin biopsy as they often cause small fiber injury prior to the onset of large fiber injury [24].

●Postherpetic neuralgia may be assessed with skin biopsy. Reduced IENFD is observed in the symptomatic region and on the opposite side to a lesser degree [48].

●Proximal mononeuropathies, such as meralgia paresthetica, is sometimes evaluated by skin biopsy when electrodiagnostic studies are nondiagnostic [49].

●Anhidrosis may be evaluated by skin biopsy to document the status of sweat gland innervation (picture 5).

●Orthostatic hypotension may be caused my multiple conditions. Immunostaining for phosphorylated alpha-synuclein at serine 129 can help to distinguish the clinically similar Parkinson's disease with orthostatic hypotension from multisystem atrophy parkinsonian type [50].

RESEARCH APPLICATIONS — Because skin biopsy is well tolerated and may be repeated multiple times, it is a powerful research tool. Intraepidermal nerve fiber density (IENFD) measurement provides a unique quantitative measure of small unmyelinated sensory fiber function that is reproducible. It has been used as an endpoint measure in peripheral neuropathy treatment trials [51-53].

Data from a study of patients with impaired glucose tolerance and neuropathy suggest that serial skin biopsies are more sensitive than other peripheral nerve measures for detecting reinnervation [54].

Various techniques to induce cutaneous denervation allow for the study of reinnervation in both normal and pathologic conditions. The most promising technique involves application of capsaicin to denervate the skin. Serial biopsies may be performed in the denervated region to study rate of reinnervation, which occurs over several months. Subjects with diabetes have a reduced rate of reinnervation [55]. This model holds promise as a means of efficiently studying potential therapeutic agents because relatively few patients can be studied for a short period of time as opposed to longitudinal clinical trials, which require large numbers of patients followed over one to two years.

Skin biopsy is also a useful method to study myelinated nerve function. It is possible to demonstrate abnormal myelin protein expression in dermal nerve fibers using immunohistochemistry. Ultrastructural and immunohistochemical analyses revealed morphologic abnormalities, many of which were similar to those observed in sural nerve biopsy tissue, but some of which were unique [56]. Analysis of dermal nerve fibers may prove useful in the study of acquired neuropathies, including inflammatory demyelinating neuropathies and diabetes [57].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Neuropathy".)

SUMMARY

●Technique – Skin biopsy is a relatively benign procedure. Most laboratories use a 3 mm punch to obtain skin tissue. This does not require a suture, and the biopsy site heals over several weeks. The biopsy site depends on the specific indication [10]. For length-dependent peripheral neuropathies, most laboratories recommend biopsy at the distal leg (8 cm proximal to the lateral malleolus) and at a more proximal site such as the lateral thigh (10 cm distal to the iliac crest). (See 'Methodology' above.)

●Interpretation – The number and morphology of axons within the epidermis are evaluated, by either qualitative or quantitative assessment, to determine intraepidermal nerve fiber density (IENFD). These values are compared with age-dependent normal values. (See 'Pathologic analysis' above.)

●Clinical applications

•Small fiber neuropathy – The most frequent clinical indication for skin biopsy in the evaluation of peripheral nerve disease is for the diagnosis of patients with suspected small fiber sensory neuropathy. In disorders affecting mainly small, unmyelinated nerve fibers, standard electrophysiologic testing is often normal, and sural nerve biopsy may be normal or only minimally abnormal. (See 'Small fiber neuropathy' above.)

•Painful sensory ganglionopathies – Skin biopsy may be particularly useful in subjects who have pain and numbness with diffuse symptoms that do not conform to a length-dependent pattern. Multiple conditions have been associated with non-length-dependent painful sensory neuropathies, including but not limited to diabetes, impaired glucose tolerance, Sjögren's disease, hepatitis C, and celiac disease. (See 'Painful sensory ganglionopathies' above.)

•Distal symmetric polyneuropathy – Skin biopsy can be used for the diagnosis of distal symmetric polyneuropathy, including those due to the involvement of small unmyelinated or large myelinated fibers when the diagnostic is uncertain after initial testing, including electrodiagnostic evaluation. (See 'Distal symmetric polyneuropathy' above.)

•Other conditions – Skin biopsy has been used to aid in the diagnosis of other neuropathic conditions including mixed fiber neuropathies, postherpetic, proximal mononeuropathies, and orthostatic hypotension. (See 'Other conditions' above.)