MUSCLE BIOPSIESMuscle and nerve biopsies can be extremely useful in the evaluation of patients with myopathies and neuropathies. That said, not everyone suspected of having a muscle or nerve disorder needs a biopsy. In this chapter, we discuss the indications and limitations for muscle and nerve biopsies, how specific muscle or nerves are selected for biopsy, and various aspects of specimen handling. Further, we review the routine stains that are performed on muscle and nerve tissue, other stains or studies that can be performed on the tissue, and when to order them. We also mention the role of skin biopsy to assess epidermal nerve fibers in the evaluation of patients with peripheral neuropathy. This chapter is not designed to make the reader a neuropathologist. However, clinicians who take care of patients with neuromuscular disease and order biopsies should have at least a working knowledge of muscle and nerve histopathology.
MUSCLE BIOPSIESMuscle biopsies are studied through a combination of various histochemistry stains on frozen sections and paraffin-embedded tissue, electron microscopy (EM), and molecular studies (e.g., enzyme assay, protein analysis by Western blot, mitochondrial DNA mutations).1–5 It is important to correlate the histopathologic findings with clinical history, neuromuscular examination, and electrodiagnostic findings.
INDICATIONS FOR MUSCLE BIOPSYA muscle biopsy may be helpful when the patient has objective muscle weakness, abnormal muscle enzymes (e.g., elevated serum creatine kinase [CK] levels), abnormal skeletal muscle magnetic resonance imaging, or myopathic electromyography (EMG) findings. These findings may point to a myopathy but not the exact etiology, and therefore a muscle biopsy may be indicated. That said, if the diagnosis is suspected on the basis of the phenotype and can be made by less invasive means, we generally opt for this first. For example, in a young boy with proximal weakness and large calves, we would first do genetic testing for a dystrophinopathy. Muscle biopsies are less helpful in evaluating patients with only myalgias, subjective weakness, or just slight elevations of CK in the absence of objective abnormalities.6
TECHNIQUESMuscle tissue can be obtained through an open (minor surgical procedure) or needle biopsy. A larger sample of tissue can be biopsied by the open surgery technique, and we prefer this method in patients who may have patchy pathology (e.g., inflammatory myopathies) or myopathies that require metabolic analysis (e.g., mitochondrial disorders or glycogen storage diseases), molecular studies (e.g., Western blotting and direct genetic analysis), or EM. Needle biopsy can also be technically difficult in patients with substantial subcutaneous tissue or whose muscles are atrophic and/or fibrotic. However, the yield of a needle biopsy can be quite high in laboratories that are experienced in handling the small amount of tissue obtained by this technique.7–10 The advantage of a needle biopsy is that it allows for the examination of multiple sites within the muscle and it is a less invasive procedure than an open muscle biopsy.
We select the specific muscle to biopsy based on the clinical examination, or occasionally based on skeletal muscle magnetic resonance imaging or EMG guidance. If the requesting physician is not the person who performs the surgery (the usual situation), communication between the two is essential to ensure that the proper site is chosen. Preferably, one should biopsy a mildly weak muscle in the Medical Research Council (MRC) grade 4/5 range to increase the yield. If the muscle is too weak (i.e., MRC grade 3 or less), the tissue typically has end-stage damage. It is often impossible to discern a myopathic process from severe neurogenic atrophy under these conditions. In patients with little, if any, weakness on examination, or those who might only have weakness in muscles that are not easily accessible to biopsy (e.g., iliopsoas muscle in a patient with only hip flexor weakness), needle EMG or skeletal muscle magnetic resonance imaging are used to select the muscle to biopsy. However, it is important to biopsy the contralateral muscle to the needle examination in order to avoid artifact from needle EMG.
We find that the easiest muscle to biopsy with open surgery is the biceps brachii and is our first choice if clinically affected. Other muscles that are commonly biopsied are the deltoid, triceps, and quadriceps. We occasionally biopsy the cervical paraspinals in patients with isolated neck extensor weakness or bent spine syndrome. The peroneus brevis muscle is useful to biopsy along with the overlying superficial peroneal nerve in patients suspected of having vasculitis. In patients with suspected distal myopathies, we have found the tibialis anterior, gastrocnemius, and forearm extensor muscles easy to biopsy. Otherwise, we tend to avoid the gastrocnemius or tibialis anterior muscle, because asymptomatic radiculopathies or unrelated axonal polyneuropathies may give a false impression that the primary abnormality is a neurogenic process and therefore overshadow an underlying myopathy.
In adults, muscle biopsies are performed under local anesthesia, but young children require sedation or general anesthesia. The biopsies are taken from the belly of the muscle, and it is important to avoid the region of the tendon. Each specimen should be about 1–2 cm in length and 0.5 cm in width. The specimens should be wrapped in slightly moist gauze and placed in separate labeled sterile containers until they reach the laboratory. It is important not to place the specimens in a container of saline else this will lead to artifact. Nor should the entire specimen be placed in fixative else the important histochemistry stains, protein/enzyme analysis, and mutation analysis cannot be performed. Again, this information needs to be communicated with the surgeon and the pathology laboratory. Because muscle disorders can be multifocal (e.g., inflammatory myopathies), we obtain at least two separate specimens, which are immediately frozen in isopentane cooled in liquid nitrogen. The frozen tissue is then sectioned and stained for routine histochemistry. In patients with prominent myalgias and tenderness, we may biopsy a piece of the overlying fascia to assess for fasciitis. Separate specimens may also be taken and again frozen immediately for biochemical analysis (e.g., for glycogen or lipid storage disorders and mitochondrial myopathies), mitochondrial DNA analysis, or for Western blot (e.g., in various forms of muscular dystrophy).
In addition, a separate piece of muscle tissue is fixed in formalin or Bouin’s fluid for paraffin sections. Paraffin sections can be particularly useful in inflammatory myopathies and vasculitis, as it allows for the examination of a somewhat larger piece of tissue than that used for histochemistry in cross section and longitudinal section and assesses inflammatory cells and vasculature more effectively (Fig. 3-1). However, due to shrinkage of the muscle tissue associated with the processing, the muscle fibers in paraffin sections are often appear cracked and are not ideal for the evaluation of histochemical abnormalities. Finally, an additional piece of muscle is usually taken for possible ultrastructural examination by EM. This small piece of muscle tissue is secured on a clamp or stretched out by suturing the muscle over a tongue blade, in order to prevent hypercontraction artifact. This tissue is fixed in glutaraldehyde for plastic (resin) embedding for EM.
Figure 3-1. Paraffin sections are useful because large, longitudinal segments of muscle fibers can be cut and stained compared to frozen sections. Marked endomysial inflammatory cell infiltrate in this biopsy of a patient with polymyositis (A). On higher power, inflammatory cell infiltrates can be seen to invade the necrotic segments (B). Paraffin sections, hematoxylin and eosin (H&E).
A standard battery of histochemical stains is used for light-microscopic evaluation of frozen sections (Table 3-1).1–5 Hematoxylin and eosin (H&E) and modified Gomori-trichrome stains assess the size, shape, and cytoarchitecture of the muscle fibers, presence of internalized nuclei, destruction of fibers (e.g., necrosis) and regeneration, as well as the supporting connective tissue (e.g., increased endomysial connective tissue as seen in dystrophies) and vasculature (vasculitis) (Figs. 3-2 and 3-3). Inflammatory cell infiltration is easily appreciated with these stains. In addition, some specific abnormalities are well demonstrated with modified Gomori-trichrome stain: ragged red fibers associated with mitochondrial myopathies (Fig. 3-4A), nemaline rods (Fig. 3-4B), tubular aggregates (Fig. 3-4C), rimmed vacuoles (Fig. 3-4D), and features of myofibrillar myopathy (Fig. 3-4E).
TABLE 3-1. MUSCLE FIBER-TYPE CHARACTERISTICS
Figure 3-2. A cluster of regenerating muscle fibers are apparent on this H&E stain.
Figure 3-3. Muscle biopsy in a patient with acute quadriplegic myopathy reveals marked atrophy and degeneration of muscle fibers on this modified Gomori-trichrome stain.
Figure 3-4. Modified Gomori-trichrome stain reveals a ragged red fiber in a patient with a mitochondrial myopathy (A), nemaline rods in a patient with congenital myopathy (B), tubular aggregates in a patient with myalgias (C), and rimmed vacuoles filled with debri in a patient with IBM (D). Myofibrillar myopathy is best recognized on the modified Gomori-trichrome stain as amorphous accumulation of dark green or bluish-purple debri and more distinct, denser cytoplasmic inclusions (E).
The myofibrillar adenosine triphosphatase (ATPase) is typically performed at three pHs, 4.3, 4.6, and 9.4, in order to assess the size and distribution of different muscle fiber types (Table 3-1 and Fig. 3-5). Individual muscle fibers can be classified into four different fiber types on the basis of their staining characteristics and physiologic properties: types 1 (slow twitch, fatigue resistant, and oxidative metabolism), 2A (fast twitch, intermediate fatigue resistance, and oxidative and glycolytic metabolism), 2B (fast twitch, poor fatigue resistance, and glycolytic metabolism), and 2C (undifferentiated and embryonic). In adults, only about 1–2% of muscle fibers are the undifferentiated type 2C fibers.11 The specific muscle fiber type is determined by the innervating motor neuron. The different muscle fiber types are normally distributed randomly, forming a so-called checkerboard pattern. Alterations in the random distribution of fiber such as seen with fiber-type grouping are a sign of denervation with subsequent reinnervation. Some myopathies are associated with a predominance or atrophy of a specific fiber type. For example, some congenital myopathies are associated with a predominance of type 1 fibers, which are also smaller in diameter than normal. Disuse and steroid myopathy are associated with preferential atrophy of type 2B fibers.
Figure 3-5. The myofibrillar adenosine triphosphatase (ATPase) is typically performed at three pHs: 4.3, 4.6, and 9.4. Type 1 fibers are lightly stained, whereas type 2 fibers are dark on ATPase 9.4 stain (A). Type 1 fibers are dark, whereas type 2 fibers are light on ATPase 4.3 stain (B). The ATPase 4.6 stains type 1 fibers dark, type 2 A fibers light, and type 2B fibers in between (C).
Periodic acid–Schiff (PAS) stain is used to assess glycogen content, which may be increased in the glycogen storage disorders (Fig. 3-6). If there is abnormal PAS staining then a PAS with diastase should be performed, as glycogen is removed with diastase but more complex carbohydrates (such as polyglucosan bodies) are resistant to digestion with diastase. Loss of some enzyme activities associated with some metabolic myopathies can be detected by specific staining protocols (e.g., myophosphorylase and phosphofructokinase). Sometimes in polyglucosan body neuropathy, PAS-positive inclusions are evident in small intramuscular nerves on muscle biopsies (Fig. 3-7). Acid phosphatase stains can highlight lysosomes that are increased in certain disorders (e.g., Pompe disease) as well as macrophages that may be present in muscle tissue (Fig. 3-8). In addition, oil red O or Sudan black can evaluate lipid content, which may be increased in patients with lipid storage myopathies (Fig. 3-9). Oxidative enzyme stains (nicotinamide adenine dinucleotide tetrazolium reductase or NADH-TR, succinate dehydrogenase or SDH, cytochrome-C oxidase or COX) are useful for identifying mitochondrial and intermyofibrillar network abnormalities (Figs. 3-10 A and B). The SDH and COX stains can be combined to highly SDH-positive, COX-negative fibers characteristic of disorders associated with mitochondrial DNA mutations (Fig. 3-10C). Target fibers and central cores are also particularly well seen with the NADH-TR stain (discussed later). In addition, a so-called trabeculated or lobulated staining pattern is seen on NADH-TR in some dystrophies, although this is not a disease-specific abnormality (Fig. 3-10D). Various stains (Congo red, crystal violet, cresyl violet, and Alcian blue) can be performed to assess for amyloid deposition (Fig. 3-11).
Figure 3-6. Scattered muscle fibers have small foci of increased glycogen deposition in subsarcolemmal regions in a patient with McArdle disease (A), periodic acid–Schiff (PAS) stain. When diastase is added to the PAS stain the abnormal accumulations are no longer evident, suggesting that the deposits were glycogen (B). PAS stain may also detect polyglucosan bodies in intramuscular nerve twigs in muscle biopsy in patients with polyglucosan body neuropathy (C).
Figure 3-7. Myophosphorylase stain demonstrates absent myophosphorylase activity in a patient with McArdle disease (A). Myophosphorylase activity in a healthy control biopsy (B). Type 2 fibers that contain more myophosphorylase stain are darker than type 1 fibers.
Figure 3-8. Acid phosphatase stains macrophages and muscle fibers lysosomes. In patients with Pompe disease, a lysosomal glycogen storage disorder, increased lysosomes are evident and brought out by acid phosphatase stain even when vacuoles may be difficult to appreciate on other routine stains such as H&E and modified Gomori trichrome.
Figure 3-9. Increased lipid droplets in muscle fibers are evident on this oil red O stain in a case of a lipid storage myopathy.
Figure 3-10. In addition to ragged red fibers seen on modified Gomori-trichrome stain (Fig. 3-4 A), mitochondrial myopathies may demonstrate muscle fibers with absent or reduced cytochrome oxidase staining (COX) (A) or increased succinic dehydrogenase staining (SDH) (B). The COX and SDH stains can be combined such that COX-negative fibers that are SDH-positive show up intensely blue (C). These stains are useful because ragged red fibers that are COX negative but SDH positive are usually associated with mitochondrial DNA mutations—though the primary mutation may still involve nuclear encoded genes that govern mitochondrial DNA. NADH-TR stain also highlights trabeculated or lobulated fibers as seen in this biopsy in a patient with muscular dystrophy (D).
Figure 3-11. Congo red stain demonstrates amyloid deposition surrounding muscle fibers and blood vessels. Under routine light microscopy, the amyloid deposition is pinkish red staining (A), apple green under polarized light, but is most easily appreciated as bright red using rhodamine optics (B).
Immunohistochemistry is important in evaluating specific types of muscular dystrophies (e.g., dystrophin staining for Duchenne and Becker muscular dystrophy; merosin and alpha-dystroglycan staining for congenital muscular dystrophy; sarcoglycans, caveolin, and dysferlin for limb girdle muscular dystrophies; and emerin for X-linked Emery-Dreifuss muscular dystrophy) (Fig. 3-12). Immunohistochemistry can also be valuable in inflammatory myopathies and vasculitis (e.g., stains for major histocompatibility antigens, complement, membrane attack complex, immunoglobulins, and appropriate inflammatory cell markers) (Fig. 3-13).
Figure 3-12. LGMD 2I. Muscle biopsies demonstrate reduced or patchy merosin staining (A), absent alpha-dystroglycan staining (B), but normal dystrophin staining (C) around the sarcolemma. Immunoperoxidase stains.
Figure 3-13. Specific types of inflammatory cells, in this case CD8+ T lymphocytes can be seen in the endomysium surrounding muscle fibers in polymyositis. Immunoperoxidase stain.
EM is used to assess the ultrastructural components of muscle fibers (e.g., the sarcolemma, sarcomeres, nuclei, and mitochondria) and vasculature (e.g., tubulofilaments in capillaries in dermatomyositis).12 Various myopathies have specific ultrastructural abnormalities that are more readily characterized by EM (e.g., nemaline rods, central cores, proliferation of abnormal appearing mitochondria, myofibrillar degeneration, vacuoles, and filamentous inclusions in nuclei and sarcoplasm) (Fig. 3-14).
Figure 3-14. Electron microscopy is useful in assessing ultrastructural abnormalities. Normal sarcomere can be appreciated with Z-disc, thick and thin filaments, glycogen granules and mitochondria (A). In critical illness myopathy, severe disruption of the sarcomere is evident with loss of the myosin thick filaments (B). Abnormal proliferation of mitochondrial with paracrystalline inclusions in this muscle biopsy of a patient with mitochondrial myopathy (C) and rods as evident in a biopsy of a patient with nemaline myopathy (D).
STRUCTURE OF NORMAL SKELETAL MUSCLESkeletal muscle is a syncytial tissue composed of sheets of individual muscle fibers with multiple nuclei. The connective tissue within muscles include the endomysium that surrounds individual muscle fibers, the perimysium that groups muscle fibers into primary and secondary bundles (fasciculi), and the epimysium that envelops single muscles or large groups of fibers. Normally, myonuclei are located adjacent to the muscle membrane (sarcolemma) and are oriented parallel to the length of the fiber. These are oval in shape and contain evenly distributed chromatin and inconspicuous nucleoli. In approximately 3% of normal adult fibers, the myonuclei lie more internal within the cytoplasm (sarcoplasm). Increased internalized nuclei are a nonspecific abnormality, as these are seen in different types of myopathies as well as in neurogenic disorders. Satellite cells are present next to the sarcolemma and are enveloped by basement membrane that surrounds the muscle fibers. Most of the sarcoplasm of the muscle fiber contains myofilaments, which form the contractile apparatus and supporting structures. Individual muscle fibers contain repeating units (sarcomeres) of interlaced, longitudinally directed thin filaments and thick filaments and perpendicularly oriented Z bands to which the thin filaments are connected (Fig. 3-14A). The sarcomere is connected to the sarcolemma via filamentous actin. The sarcolemma is composed of various protein complexes and is connected to the extracellular matrix. Greater detail of the sarcolemmal proteins and extracellular matrix is discussed in Chapter 27.
The T tubules are composed of invaginations of the sarcolemmal membrane into the interior of the muscle fibers. Their course is parallel to the Z bands and they are surrounded on each site by the sarcoplasmic reticulum. The T tubules allow for rapid depolarization of muscle membrane deep within muscle fiber cells and the accelerated release of calcium from the sarcoplasmic reticulum during excitation.
Adult muscle fibers are polygonal in appearance but are more rounded in shape in infancy and early childhood. The cross-sectional diameter of individual fibers varies depending on the specific muscle, fiber type, and age of the individual. The motor unit comprises the motor neuron and the muscle fibers it innervates. The individual muscle fibers of a motor unit are normally randomly distributed as previously mentioned, within a sector approximating 30% of the muscle’s cross-sectional diameter.
The percentages of type 1, 2A, and 2B fibers differ in various muscle groups, and it is important to be aware of the normal percentages of these fibers in the biopsied muscle for accurate assessment.13 The most commonly biopsied muscles (i.e., biceps brachii, triceps, and quadriceps) have approximately equal amounts of the three major fiber types, although the deltoid muscle has more type 1 fibers than type 2A and 2B. Because muscle fibers from a single motor unit are randomly distributed among muscle fibers of different motor units and fiber types, a checkerboard or mosaic pattern is appreciated on ATPase stains (Fig. 3-5).
Although ATPase stain is primarily used to assess fiber type, we can often ascertain the fiber types from other standard stains (Table 3-1). For example, type 1 fibers stain more intensely with modified Gomori-trichrome, lipid, and oxidative enzyme stains than type 2 fibers because of the increased number of mitochondria and oxidative metabolism associated with type 1 fibers. In contrast, type 2 fibers, which are involved with glycolytic metabolism, stain more intensely with PAS, as these contain more glycogen but are lighter staining on modified Gomori-trichrome, lipid, and oxidative enzyme stains.
The diameters of individual muscle fibers are assessed in order to characterize their size. Quantitative analysis is performed by measuring the mean and range of the diameters for each different fiber type.14–17 Importantly, the diameters of muscle fibers increase to a point during childhood until the early teens. At 1 year of age the mean muscle fiber diameter is approximately 16 μm. The size increases by about 2 μm/yr until the age of 5 years and subsequently by 3 μm/yr until 9 years of age. By 10 years of age, mean muscle diameters range from 38 to 42 μm. Normal adult size is reached between the ages of 12 and 15 years.17 There is usually less than 12% difference in the largest mean fiber diameters between the major fiber types. Both types 1 and 2 adult muscle fibers are larger in men than in women. Type 2 fibers are usually larger than type 1 fibers in men; type 1 fibers are larger than type 2 fibers in women. The diameter of muscle fibers is also dependent on the specific muscle biopsied. For example, in adults, the diameters of muscle fibers in the biceps brachii are as follows: type 1 fibers 64.3 +/- 3.7 μm and type 2 fibers 72.7 +/- 5.3 μm in men and type 1 fibers 56.8 +/- 4.8 μm and type 2 fibers 54.6 +/- 7.0 μm in women. In the vastus lateralis, the diameters of muscle fibers are slightly different: type 1 fibers 59.5 +/- 6.4 μm and type 2 fibers 64.8 +/- 8.1 μm in men and type 1 fibers 58.8 +/- 6.1 μm and type 2 fibers 49.9 +/- 6.2 μm in women.14
REACTIONS TO INJURYMuscle abnormalities may be classified on histopathologic and etiologic grounds into three major categories: (1) neurogenic atrophy: a pattern of muscle pathology consequent to denervation, and if present, reinnervation; (2) myopathies: inherited and acquired diseases characterized by abnormalities in the muscle fiber itself; these include dystrophies, congenital, inflammatory, metabolic, and toxic myopathies; and (3) disorders of the neuromuscular junction. Patients with neuromuscular junction defects usually have only slight and nonspecific alterations apparent on routine light microscopy and are rarely biopsied except at very specialized centers.1–5
Upon review of muscle biopsy slides, specific features on various stains are important to note. It is essential to assess the size and variability of muscle fibers, the distribution of fiber types, the size and location of the myonuclei, the presence of necrotic and regenerating muscle fibers, other alterations in the cytoarchitecture and organelles (e.g., the presence of target fibers, cores, vacuoles, tubular aggregates, and ragged red fibers), and any abnormal accumulation of glycogen or lipid. Besides the muscle fibers, we evaluate the surrounding vasculature (is there evidence of vasculitis and thickened basement membranes?) and the supportive tissue (is there increased endomysial connective tissue, edema, or amyloidosis?). One should characterize any inflammatory cell infiltrate making note of the type (lymphocytes, plasma cells, eosinophils, and macrophages), the location (endomysial, perimysial, and perivascular), and if there is cellular invasion of nonnecrotic or just necrotic appearing fibers. We discuss some of the common abnormalities seen on muscle biopsy in the following section, but in more detail in the subsequent chapters where specific disorders and their characteristic histologic features are described.
In the setting of axonal degeneration, the muscle fibers within that motor unit lose their neural input and undergo denervation atrophy. This leads to decreased synthesis of myofilaments, degeneration of myofibrils, and a reduction in the size of the muscle fiber.18 The atrophic fibers lose their polygonal appearance and look angulated in shape (Fig. 3-15). Neurogenic disorders affect motor nerves that innervate both type 1 and 2 fibers. Therefore, in early denervation, muscle biopsies reveal scattered, atrophic angulated muscle fibers of both fiber types. As more motor nerves degenerate, rather than seeing isolated atrophic angulated fibers, there are groups of adjacent muscle fibers that are atrophic (grouped atrophy). A feature of denervation is the presence of the so-called target fibers. Reorganization of the cytoarchitecture within muscle cells results in a rounded central zone of disorganized filaments that contain fewer mitochondrial and glycogen. Target fibers have three zones that are circumferentially oriented, which are best seen on NADH-TR staining (Figs. 3-16A and B). The innermost zone is devoid of mitochondrial, glycogen, phosphorylase, and ATPase enzymatic activity; the second zone has increased enzymatic activity, whereas the third zone exhibits intermediate enzymatic activity. Target fibers occur in neurogenic disorders in the course of reinnervation and need to be distinguished from central cores in which there are just two zones of staining (Fig. 3-16C). Central cores are specific for the congenital myopathy central core disease. The so-called moth-eaten or targetoid fibers resemble targets and cores on the NADH-TR stain but have less circumscribed patches of reduced oxidative enzyme staining again without a distinct intermediate zone of enzyme activity. Moth-eaten targetoid fibers (Fig. 3-16D) are nonspecific and can be seen in myopathic and neurogenic disorders. Both central cores and target fibers preferentially affect type 1 fibers. In contrast to central core myopathy in which the cores are present in the majority of type 1 fibers, the percentages of fibers with target and targetoid abnormalities are less abundant. Target fibers and central cores can also be appreciated on other stains such as the ATPase, PAS, and modified Gomori-trichrome stains (Fig. 3-16E).
Figure 3-15. Neurogenic atrophy. Denervation results in muscle fibers becoming atrophic and angulated in appearance (A). Several atrophic and angulated fibers clustered together are referred to as group atrophy (B). If surrounding nerve fibers sprout and reinnervate nearby denervated muscle fibers, the newly reinnervated fibers assume the fiber type of the motor nerve that now innervates them. This leads to the loss of the mosaic pattern on ATPase stains and the appearance of fiber-type grouping (C). ATPase 4.3.
Figure 3-16. In the course of reinnervation, target fibers may develop. True target fibers have three zones in the center of the muscle fibers that are best seen on NADH-TR staining, at low power (A), and at higher power (B). The innermost zone is pale; the second zone has increased enzymatic activity, whereas the third zone exhibits intermediate enzymatic activity. Central cores resemble targets, but there is not a second zone with increased enzyme activity (C). In targetoid or the so-called moth-eaten fibers the zones of reduced activity are even less distinct (D). Target fibers and cores can also be appreciated on other stains such as the modified Gomori-trichrome stain (E). On the Gomori-trichrome stain, the center of the target fibers stain dark and are surround by pale staining zone.
Denervated muscle fibers send out trophic signals that lead nearby unaffected axons to sprout collateral branches, in an attempt to reinnervate the newly denervated muscle fibers. Once successful reinnervation is accomplished, the newly reinnervated muscle fiber assumes the physiologic properties of the reinnervating neuron. This may lead a type 1 fiber to become a type 2A fiber or a type 2B fiber to become a type 1 fiber. As a consequence, the normal checkerboard appearance of muscle tissue is replaced by large groups of single muscle fibers, all with the same fiber type (e.g., fiber-type grouping) (Fig. 3-15C). If these larger motor nerves subsequently degenerate, large areas of atrophic fibers of the same fiber type are seen—a different type of grouped atrophy.
In contrast to neurogenic atrophy, myopathic disorders are associated with a wider spectrum of histopathologic alterations (Fig. 3-17). Remember that muscle is a syncytium formed from the fusion of thousands of myoblasts. Because of its syncytial nature, histopathologic abnormalities may be focal rather than occurring along the entire length of a muscle fiber (e.g., segmental necrosis). Genetic disorders can manifest discrete abnormalities, with other regions of the single fiber appearing relatively normal. An example of this can be seen in mitochondrial myopathies in which the histopathologic alterations are dependent on the degree of abnormal mitochondria, which in turn is a reflection of the percentage of mutated mitochondrial DNA in the region. Thus, when cut longitudinally, one may appreciate segments of the muscle fiber with a ragged red appearance, which do not stain with cytochrome oxidase, whereas other nearby segments of the same fiber may be normal. In dystrophies, one often sees scattered necrotic muscle fibers on the cross section. However, if the tissue is cut longitudinally, one sees that necrosis is segmental in nature. Likewise, inflammatory myopathies are multifocal, resulting in infiltrates surrounding and invading segments of muscle fibers along their length.
Figure 3-17. Variability in muscle fiber size, increased internalized nuclei, muscle fiber splitting, and small intracytoplasmic vacuoles are nonspecific myopathy features appreciated on this modified Gomori-trichrome stain.
Myopathies are usually associated with a random loss of muscle fibers belonging to different motor units. Atrophy of muscle fibers is a common histopathologic feature in myopathies, but rather than fibers becoming angular as in neurogenic atrophy, these usually become more rounded in appearance in myopathic disorders. Small groups of atrophic fibers of similar type may be observed in myopathies due to muscle fiber splitting, degeneration, and regeneration; however, large areas of group atrophy or fiber-type grouping are more typical of neurogenic atrophy. Preferential atrophy or hypotrophy of type 1 fibers is seen in certain myopathic disorders (e.g., myotonic dystrophy and various congenital myopathies). On the other hand, preferential type 2 fiber atrophy can be seen in certain endocrine disorders (e.g., steroid myopathy), as well as a complication of disuse.
Besides atrophy of muscle fibers, hypertrophy can develop in response to increased load, either in the setting of exercise or in pathologic conditions where other muscle fibers are injured. Large fibers may divide along a segment (muscle fiber splitting) so that, in cross section, a single large fiber contains a cell membrane traversing its diameter. Because both chronic myopathic and neurogenic disorders can be associated with a mixture of atrophic and hypertrophic fibers, increased variability of muscle fiber size is a nonspecific abnormality.
Necrosis is a feature more common in myopathies, but it can also occur in denervated muscle fibers (Fig. 3-18). A single muscle fiber can undergo either total necrosis or segmental necrosis, but again, given the syncytial nature of muscle, atrophy along the entire fiber length is rare. The more common form of muscle tissue loss is referred to as segmental necrosis in which a relatively small segment of the single muscle fiber is affected. The site of necrosis may be focal at first, but it extends longitudinally along the muscle fiber with disease progression. Segmental necrosis is best appreciated on paraffin or semithin sections of muscle fibers cut longitudinally. With segmental necrosis, the affected portion of the single muscle fiber becomes more rounded and the sarcoplasm begins to have a featureless ground-glass appearance. Semithin and EM sections reveal degeneration of the Z disc and myofibrillar network as well as abnormal mitochondria. Macrophages are recruited into the area and infiltrate the necrotic segments in order to digest the disintegrating muscle tissue and damaged tissue. In certain diseases (polymyositis and inclusion body myositis [IBM]), macrophages and lymphocytes may invade nonnecrotic tissue such that a muscle fiber can be “severed” into distinct segments.
Figure 3-18. A necrotic muscle fiber is pale staining in comparison to surrounding muscle fibers in cross section, H&E stain. (A) Segmental necrosis is often well appreciated on paraffin sections in which large, longitudinal segments can be visualized. The striations of the sarcomeres can be appreciated in normal fibers, whereas the necrotic segment of an adjacent fiber loses the striations. The necrotic segment is invaded by macrophages (B). Paraffin section, H&E stain.
Repair of necrotic segments can occur and begins with the proliferation of adjacent satellite cells in the region of the destroyed portion of the fiber.19 The satellite cells align next to each other to form myotubes. Several myotubes form per segment and adhere to the surrounding basal lamina. The expansion of myotubes occurs laterally and longitudinally, eventually reaching and fusing with the healthy muscle tissue stumps. The regenerating muscle fibers can be appreciated by their large internalized nuclei with prominent nucleoli, and their basophilic cytoplasm that is laden with ribonucleic acid (RNA) (Fig. 3-19). Old damage can be ascertained by the increase in the number of internalized nuclei (Fig. 3-17). Myonuclei, which usually lie along the subsarcolemmal membrane, are more internalized in regenerated segments.
Figure 3-19. Regenerating muscle fibers are smaller and more basophilic than normal fibers and contain enlarged nuclei sometimes with nucleoli, as these are very active in trying to replenish necessary muscle proteins. H&E stain.
Other characteristics of myopathic injury include alterations in structural proteins or organelles, formation of vacuoles, and accumulation of intracytoplasmic deposits. Increased endomysial connective tissue is a common feature of muscular dystrophies but is also seen in chronic inflammatory myopathies and severe end-stage neurogenic atrophy. One of the most common reasons for a muscle biopsy in adults is to assist in diagnosis of a primary inflammatory myopathy. The characteristic histopathologic features on muscle biopsies in dermatomyositis are perifascicular atrophy and perivascular, perimysial inflammatory cell infiltrate with many plasmacytoid dendritic cells (Fig. 3-20A). On the other hand, polymyositis is associated with endomysial T cells that surround and often appear to invade nonnecrotic muscle fibers (Fig. 3-20B). IBM shares these features with polymyositis, but in addition there is often rimmed vacuoles and various inclusions apparent on light microscopy and EM (Fig. 3-20C). It is important to note that rimmed vacuoles and inclusions are not seen in at least 20% of any given IBM biopsy so the diagnosis of IBM cannot be excluded in the absence of these findings. Furthermore, one will not see rimmed vacuoles on paraffin sections, only on frozen sections—so it is imperative to do histochemistry staining of frozen section and not just paraffin sections. Immune-mediated necrotizing myopathy is associated with scattered necrotic and regenerating muscle fibers in the absence of significant inflammatory cell infiltrate (Fig. 3-18). Less common forms of inflammatory myopathy include granulomatous or giant cell myositis (Fig. 3-20D) and eosinophilic myositis (Fig. 3-20E). A precautionary note is that inflammatory cell infiltrates are seen in dystrophies and other types of myopathy and thus are not specific for an immune-mediated process.
Figure 3-20. The characteristic feature of dermatomyositis is perifascicular atrophy and perivascular, perimysial inflammation (A), H&E stain. Polymyositis is associated with endomysial inflammatory cell infiltrates surrounding and often appearing to invade nonnecrotic muscle fibers (B), H&E stain. Inclusion body myositis likewise has features of polymyositis but rimmed vacuoles are often apparent (C), H&E stain. In sarcoidosis and granuolomatous myositis the biopsies reveal granulomas (D), H&E stain. Eosinophils are prominent among the inflammatory cells in eosinophilic myositis, but these cells may rarely also seen in inflammatory dystrophies (E), H&E stain.
NERVE BIOPSIESAs is true for muscle biopsies, the interpretation of a nerve biopsy requires correlation of histologic changes, with clinical information including the results of electrophysiologic investigations. Nerve biopsies are generally less useful than muscle biopsies because the pathologic abnormalities are often nonspecific and frequently do not help distinguish one form of peripheral neuropathy from the other.20–23 In addition, there is increased morbidity associated with the removal of a segment of sensory nerve, which leads to permanent numbness in the corresponding cutaneous distribution. Also, nerve biopsies can be complicated by significant neuropathic pain in the distribution of the nerve for several months and the potential for growth of painful neuromas. Therefore, we do not recommend nerve biopsies just because the patient has a generalized neuropathy of undetermined etiology despite an extensive laboratory evaluation. This situation is quite common, as discussed in Chapter 22.
INDICATIONS FOR NERVE BIOPSYSuspected amyloidosis and vasculitis are the major indications for nerve biopsy. Amyloidosis should be considered in patients with a monoclonal gammopathy, autonomic neuropathy, systemic signs of amyloidosis (e.g., renal insufficiency or cardiomyopathy), or a family history of amyloidosis. Vasculitic neuropathy is in the differential diagnosis in people presenting with a history of multiple mononeuropathies, particularly when of acute onset and painful, and when there is an underlying connective tissue disease (e.g., systemic lupus erythematosus and rheumatoid arthritis), eosinophilia or late-onset asthma (Churg–Strauss syndrome), renal failure or chronic sinusitis, hepatitis B or C, an elevated erythrocyte sedimentation rate, or antinuclear cytoplasmic antibody. Additional indications for nerve biopsy include other autoimmune inflammatory conditions (e.g., sarcoidosis), possible infectious processes (e.g., leprosy), and tumor infiltration (e.g., lymphoma and leukemia). Also, a nerve biopsy may be required for the diagnosis of a tumor of the peripheral nerve (e.g., perineurioma). Less commonly, nerve biopsy may be warranted to diagnose uncommon forms of hereditary neuropathy when DNA testing is not available or is negative (e.g., giant axonal neuropathy and polyglucosan body neuropathy).
TECHNIQUESWe usually biopsy a superficial sensory nerve that is clinically affected and also abnormal on sensory nerve conduction studies. The most common nerve biopsied is the sural nerve. We prefer to biopsy the sural nerve in the mid-shin approximately one-third to one-fourth of the distance from ankle to knee, as opposed to the lateral ankle itself where the nerve may be more prone to trauma and healing may not be as good (Fig. 3-21). Patients should be warned that following the sural nerve biopsy, there is often pain for several months as well permanent loss of sensation on the lateral aspect of the ankle and foot.20 A superficial peroneal nerve biopsy is particularly useful when vasculitic neuropathy is suspected and there is foot drop, because the underlying peroneus brevis muscle can also be biopsied through the same incision site, thereby increasing the diagnostic yield (Fig. 3-22). Biopsy of the superficial peroneal nerve will lead to numbness of the dorsum of the foot and again often neuropathic pain for several months.20 If only the upper extremities are involved, the superficial radial nerve can be biopsied; however, this leads to numbness of the dorsum of the hand, which is problematic for most patients. Importantly, because of sampling error, the single small segment of distal sensory nerve may not be representative of focal disease processes elsewhere in the peripheral nervous system, especially in processes with predominant motor involvement. On rare occasions when a patient has a multifocal process and the lesion appears proximal (e.g., amyloidomas, inflammatory process, and tumors), a fascicular nerve biopsy of a lesion in the root, plexus, or proximal nerve may be required. This procedure should only be performed, however, by neurosurgeons experienced in the technique and where the tissue can be processed appropriately in the neuropathology laboratory.
Figure 3-21. The sural nerve is usually biopsied approximately one-third up from the ankle just lateral to the midline in the grove made by the Achilles’ tendon. It is important for the surgeon to isolate and distinguish the saphenous vein from the sural nerve as they lie next to each other. The saphenous vein can look nearly identical to the sural nerve, often leading to an erroneous “nerve biopsy” with a lumen if care is not taken. A silk suture is gently lifting the sural nerve away from the saphenous vein (A). The nerve is injected proximally with lidocaine (B), and then is dissected away from the surrounding tissue (C). A 4-cm segment is biopsied and divided into separate specimens for frozen section, paraffin embedding, semithin, EM, and teased fiber preparations (D).
Figure 3-22. A combined superficial peroneal nerve and muscle biopsy is useful when looking for vasculitis. The nerve is typically found between one-third and one-fourth up from the lateral malleolus and approximately 1–1.5 cm anterior to the fibula. The nerve in this position can lie above or beneath the fascia overlying the peroneus brevis muscle, so both can be taken from a single incision. (Modified with permission from Mendell JR, Erdem S, Agamonolis DR. Peripheral nerve and skin biopsies. In: Mendell JR, Kissel JT, Cornblath DR, eds. Diagnosis and Management of Peripheral Nerve Disorders. New York, NY: Oxford University Press; 2001.)
Nerve biopsies are performed under local anesthesia in adults; general anesthesia is often required to obtain an adequate specimen from children or when a proximal nerve segment needs to be biopsied (e.g., root, plexus, or proximal nerve). The pathology laboratory should be contacted in advance of the surgery so that the tissue can be picked up directly from the operating room and processed immediately. Local anesthetic should be injected into the nerve just proximal to the site of transection in awake patients in order to reduce pain associated with sectioning the nerve (Fig. 3-21B). A 4–5-cm long section of nerve should be excised. The specimen can be wrapped in a saline-moistened gauze (not drenching wet).
The nerve biopsy is divided into several portions so that different types of studies can be performed (Fig. 3-21E). We generally take a small piece at the most proximal end for the frozen section. This piece is rapidly frozen in mounting medium for immunofluorescence studies. These studies can reveal the deposition of immunoglobulins or other inflammatory markers, such as complement or fibrinogen. Routine paraffin embedding (following fixation in formalin) is performed on a portion of tissue taken from the proximal and distal segments of the nerve biopsy (approximately 1 cm in length at both ends). The paraffin sections can be stained with H&E, trichrome, Luxol fast blue (stains myelin blue), Bodian stain or neurofilament stains for axons Fig. 3-23).3,21,24–27 Congo red, Alcian blue, or cresyl violet should be done to look for amyloid (Fig. 3-24). A PAS stain is useful when polyglucosan body neuropathy is a consideration (Fig. 3-25, and a Fite stain can be done to look for the acid-fast bacilli, if lepromatous neuropathy is possible (Fig. 3-26). Immunohistochemistry studies can be done to better assess inflammatory cell infiltrates (Fig. 3-27), and other specific stains can be done to better evaluate Schwann cells and perineurial cells when indicated. For example, immunoreactivity against the Schwann cell marker S-100 is useful for schwannomas and neurofibromas (Fig. 3-28), whereas immunoreactivity to epithelial membrane antigen (EMA), which is present on perineurial cells, is helpful in diagnosing perineuriomas. The paraffin-embedded tissue is most useful for evaluating signs of vasculitis, other inflammatory cell infiltrates including granulomas and lymphoma, infection (e.g., leprosy), and amyloidosis. Because the pathology can be multifocal, we like to take sections for paraffin embedding at the proximal and distal ends of the biopsy segment (Fig. 3-21E). In addition, loss of myelinated nerve fibers can be appreciated with various stains of paraffin-embedded tissue.
Figure 3-23. Paraffin sections of nerve biopsy. Myelin stains pink on modified Gomori trichrome in this normal nerve seen in cross section (A) and longitudinally (B). A reduction in myelinated fibers is apparent by the loss of pink stain on this modified Gomoritrichrome stain (C) and as blue staining myelinated nerve fibers on a Luxal fast blue stain (D); however, it is not possible to tell if this is due to primary demyelinating neuropathy or secondary demyelination from a primary axonopathy. SMI-31 stains phosphorylated neurofilaments that are abundant in normal axons, as seen in this normal nerve (E). H&E stain does not distinguish myelinated axons very well, but is useful to look for vasculitis and other inflammatory cell infiltrates, as seen in this biopsy of a patient with lymphoma (F).
Figure 3-24. Familial amyloid polyneuropathy. Nerve biopsy demonstrates abnormal accumulation of amyloidogenic material in the endoneurium in the biopsy of a patient with a transthyretin mutation. The material stains faintly pink on H&E (A). With Congo red under routine light microscopy, amyloid stains intensely red when viewed under rhodamine optics (B). Amyloid stains blue with Alcian blue stain (C).
Figure 3-25. PAS stain demonstrates polyglucosan bodies within the axons in polyglucosan body neuropathy, as seen in cross section (A) and longitudinal section (B).
Figure 3-26. Borderline leprosy. Nerve biopsy in a patient with leprosy reveals red staining bacilli using the Fite stain on paraffin sections.
Figure 3-27. Immunoperoxidase stain reveals perivascular CD3+ T lymphocytes in a nerve biopsy in a patient with chronic inflammatory demyelinating polyneuropathy (CIDP).
Figure 3-28. Neurofibroma. The nerve fascicle has a lobulated appearance, H&E stain (A). The cells have wavy, elongated nuclei, and the background material is loosely arranged and myxoid. Bands of thick collagen are apparent in the center of the tumor. Some of the proliferating tumor cells are immunoreactive for S-100, suggesting Schwann cell origin (B).
The remainder of the tissue is stretched delicately on a tongue blade or kept isometric with sutures and fixed in glutaraldehyde or other fixatives (e.g., Karnovsky’s fixative). Some of this tissue will then be embedded in plastic and processed for toluidine blue-stained semithin sections (10 μm) and thin sections (1 μm) for EM.3,21,24–28 The semithin and EM analyses are most important in assessing the axons, Schwann cells, and myelin sheath of myelinated nerve fibers as well as in looking at abnormalities in small unmyelinated nerve fibers (Fig. 3-29). Quantitative morphometric methods can be employed to assess numbers of individual large or small myelinated and unmyelinated fibers in the biopsy, as certain neuropathies have a predilection for certain nerve fiber types. However, this is not routinely done as it is time consuming and often of limited value. Other portions of this fixed material may be used for teased nerve fiber analysis (Fig. 3-30). With this method, individual myelinated fibers are separated from the nerve fascicles and lightly stained, allowing examination of the integrity and thickness of the myelin sheath as well as revealing alterations in internode length. Thus, one can better quantify the degree of demyelinated or thinly myelinated axon, axons with increased or redundant myelin, and axons undergoing active Wallerian degeneration. Teased fiber preparation, however, is very labor intensive and often does not add much to what can be assessed from the semithin and EM sections; thus, it is reserved for more difficult cases (e.g., question of CIDP in biopsy with mild or nonspecific abnormalities on semithin or EM).
Figure 3-29. Semithin section reveals a normal nerve fascicle.
Figure 3-30. Teased nerve fibers. A normal teased fiber internode is seen (A) as well as a short, demyelinated internode (B). A teased nerve fiber segment undergoing Wallerian degeneration with myelin ovoids is appreciated in (C). Redundant folds of myelin lead to formation of tomacula (Latin for sausage) that are best appreciated on teased fiber preparations (D) and are commonly seen in hereditary neuropathy with liability to pressure palsies and occasionally in other forms of Charcot–Marie–Tooth disease.
STRUCTURE OF NORMAL NERVEPeripheral nerves are composed of axons, Schwann cells, myelin sheaths, and supporting tissue. Individual nerve fibers are surrounded by endoneurial connective tissue and grouped into fascicles encased by perineurial sheaths. All the fascicles within a nerve in turn are surrounded by epineurial connective tissue. A blood–nerve barrier is created between the perineurial cells and endoneurial capillaries derived from the vasa nervorum, both of which form tight junctions. The blood–nerve barrier appears to be relatively less competent within nerve roots, dorsal root ganglia, autonomic ganglia, and terminal twigs. The nerve–CSF barrier is formed by the tight junctions between the cells that form the outer layer of the arachnoid membrane. These cells fuse with the perineurium of the roots and cranial nerves as these leave the subarachnoid space.
Myelinated and unmyelinated nerve fibers intermingle within each fascicle. Further, along the course of the entire nerve, individual nerve fibers course in and out of different fascicles. In the sural nerve, which is most commonly biopsied, myelinated fibers range between 2 and 15 μm in diameter and have a bimodal distribution. There are approximately twice as many small myelinated axons as there are large myelinated fibers. Segments of myelinated fibers (internodes) are separated by nodes of Ranvier. A single Schwann cell supplies the myelin sheath for each internode. The thickness of the myelin sheath is directly proportional to the diameter of the axon, and the larger the axon diameter, the longer the internodal distance. The ratio of the diameter of the axon to the diameter of the entire nerve fiber (axon plus its surrounding myelin) or G-ratio is approximately 0.6. A higher-than-normal diameter ratio implies that the axons are thinly myelinated. In contrast, lower G ratios are seen in axonopathies with axonal atrophy or rare conductions with redundant myelin (tomaculous neuropathy). Unmyelinated axons are more numerous than myelinated axons and range in diameter from 0.2 to 3 μm. Anywhere from 5 to 20 unmyelinated axons are enveloped by a single Schwann cell.
Schwann cells, regardless of their association with myelinated or unmyelinated fibers, have pale oval nuclei with an even chromatin distribution and an elongated bipolar cell body. On EM, Schwann cells can be differentiated from fibroblasts because Schwann cells have a basement membrane. Within axons there are various organelles and cytoskeletal structures, including mitochondria, vesicles, smooth endoplasmic reticulum, lysosomes, neurofilaments, and microtubules. Because protein synthesis occurs in the cell body rather than the axon itself, essential proteins and other substances synthesized in the perikaryon are transported down the axon via axoplasmic flow. A retrograde transport system serves as a feedback to the cell body. These transport systems are dependent on the microtubules and neurofilaments as well as specific proteins such as dynein and dynactin within the axons. At the distal nerve terminal, dense-cored and coated vesicles are found.
REACTIONS TO INJURYAlthough disease processes affecting nerves have different pathogenic mechanisms, these lead to two principal reactions to injury: demyelination and axonal degeneration.3,21,24–28 Damage to Schwann cells or the myelin sheath itself can lead to demyelination. Because these diseases affect individual Schwann cells to varying degrees, the process is characteristically segmental along the length of the nerve. The disintegrating myelin is phagocytosed by Schwann cells and macrophages. Schwann cells are also stimulated to remyelinate the denuded axon. These newly remyelinated axons are thinner in total diameter and the internodes are shorter than normal—features that are best seen with teased nerve preparations. However, one can also appreciate the thinly myelinated axons on semithin sections and on EM (diameter ratio greater than 0.6). With sequential episodes of demyelination and attempted remyelination, concentric tiers of Schwann cell processes accumulate around the axons forming the so-called “onion bulbs” (Fig. 3-31). Some disease processes are associated with inclusions within Schwann cells (e.g., metachromatic leukodystrophy and certain toxic neuropathies). Other abnormalities in the myelin sheath include tomaculae (redundant folds of myelin characteristic of hereditary neuropathy with liability to pressure palsies or HNPP) and widened periodicity of compacted myelin (seen in neuropathy associated with myelin-associated glycoprotein antibodies).
Figure 3-31. Onion bulb formation. With recurrent bouts of demyelination and remyelination, concentric layers of Schwann cell processes accumulate around the axons forming onion bulbs. Prominent onion bulbs can be seen in chronic inflammatory demyelinating neuropathy as in this case but are more typical of hereditary demyelinating neuropathies (i.e., Charcot–Marie–Tooth disease types 1, 3, and 4) on semithin section (A) and on electron microscopy (B).
Primary damage to the axon may either be due to a discrete, localized event (trauma, ischemia, etc.) or be due to an underlying abnormality of the neuronal cell body or ganglion (neuronopathy) or its axon (axonopathy). These processes lead to axonal degeneration with secondary disintegration of its myelin sheath (Fig. 3-32A). If a nerve is transected, the distal portion of the nerve undergoes an acute disintegration (termed Wallerian degeneration) characterized by breakdown of the axon and its myelin sheath into fragments forming small oval compartments (i.e., myelin ovoids). These breakdown products undergo phagocytosis by macrophages and Schwann cells. Most neuronopathies or axonopathies evolve more slowly; therefore, evidence of active axon and myelin breakdown is scant because only a few fibers are degenerating at any given time. The proximal stumps of axons that have degenerated may sprout new axons that attempt to grow along the course of the degenerated axon. Small clusters of these regenerated axons, which are small in diameter and thinly myelinated, can be recognized in cross section on semithin and EM sections (Fig. 3-32B). Also, as axonal transport of essential proteins and other substances synthesized in the perikaryon is often impaired in axonopathies, axonal atrophy becomes apparent on the semithin and EM sections (G ratio less than 0.6). In contrast, enlarged axons are seen in giant axonal neuropathy and hexacarbon toxicity. Polyglucosan bodies are appreciated on semithin sections (Fig. 3-32C). These are nonspecific, and although rare polyglucosan bodies may be seen on nerve biopsies in elderly and in diabetics, they are much more abundant in patients with adult polyglucosan body disease.
Figure 3-32. A semithin section reveals several fibers undergoing active axonal degeneration (Wallerian degeneration) (A). As nerve fibers attempt to regenerate they send out nerve sprouts. These can be appreciated as groups of thinly myelinated nerve fibers surrounded by the same basement membrane (B). Polyglucosan bodies are abundant in nerve biopsies in patients with adult polyglucosan body disease. They appear as round, thinly lamellated inclusions within axons (C).
In addition, nerve biopsies can reveal evidence of disease processes similar to those found in other organ systems. Amyloid deposition around blood vessels or within the endoneurium can be seen in systemic amyloidosis or in a familial amyloidotic polyneuropathy (Fig. 3-24). In systemic or isolated peripheral nerve vasculitis, there is transmural infiltration of vessel walls by inflammatory cells associated with fibrinoid necrosis of the vessel walls (Fig. 3-33). Because nerve fibers course between different fascicles along the length of the nerve, patchy asymmetric loss of axons within and between fascicles is a characteristic finding of ischemic nerve injury as seen in vasculitis. Infiltration of the nerve by neoplastic or inflammatory cells can also be recognized. Leprosy is one of the most common etiologies of polyneuropathy in the world. When granulomas or diffuse inflammation of the nerve is seen, a Fite stain can be done to look for the acid-fast bacilli (Fig. 3-26).
Figure 3-33. Nerve biopsy in a patient with Churg–Strauss syndrome reveals necrotizing vasculitis. Paraffin section, H&E stain.
SKIN BIOPSYSkin biopsies are increasingly being performed to evaluate patients with peripheral neuropathy.20,29–41 These are most useful in patients with small fiber neuropathies in which other testing modalities provide normal or inconclusive results. Because nerve conduction studies only assess the conduction of large myelinated nerve fiber, patients with pure small fiber neuropathies will have normal nerve conduction studies. In at least a third of people with painful sensory neuropathies, intraepidermal nerve fibers density on skin biopsies represent the only objective abnormality present following extensive evaluation.35
The rationale behind performing most skin biopsies is to measure the density and assess the morphology of intraepidermal nerve fibers. These fibers represent the terminals of Aδ and C nociceptors, and these may be decreased in patients with small fiber neuropathies in whom nerve conduction studies and routine nerve biopsies are often normal. Skin biopsies are relatively easy to perform and are associated with a much lower risk than standard nerve biopsies. However, there are several drawbacks to skin biopsies. Importantly, these usually just confirm what you already know about the patient. That is, if a person complains of symmetric burning or tingling pain in the distal lower extremities, has normal strength and deep tendon reflexes, and has normal nerve conduction studies, then he or she likely has a small fiber neuropathy. Skin biopsies are often not useful in identifying the etiology of the neuropathy. The exception is biopsy of skin lesions in suspected cases of lepromatous neuropathy (Fig. 3-34). As stated in the previous section on nerve biopsies, we generally do not do a biopsy in order to prove that a patient has a neuropathy; rather we do so in order to identify the etiology, hopefully a treatable one. That said, assessing intraepidermal nerve fiber density and morphology may play a role in the future by defining the natural history of various neuropathies, monitoring response of the neuropathy to various therapies, and screening for the development of toxic neuropathies (e.g., during chemotherapy).35 In addition, skin biopsies may be done to confirm a diagnosis of dermatomyositis.
Figure 3-34. Borderline leprosy. Skin biopsy demonstrates marked inflammatory cell infiltrate, H&E (A). Red staining bacilli are evident on higher power with a Fite stain (B)
Skin biopsies are usually done by performing a 3-mm punch biopsy of the skin under local anesthesia in the lower leg in an affected region. Other regions can be sampled to assess if there is a length-dependent loss of intraepidermal nerve fibers (e.g., in the dorsum of the foot, thigh, or forearm). The tissue is fixed in formalin, and then immunostaining protein gene product 9.5 (PGP 9.5) is applied to demonstrate the small intraepidermal fibers (Fig. 3-35). Morphometric methods are used to assess the number and complexity of these nerves, through parameters such as the linear density (number of fibers per millimeter of biopsy) or total length of intraepidermal nerve fibers. The morphology of the intraepidermal nerve fibers can also be assessed. Axonal swellings may be an early marker of small fiber neuropathy and may be appreciated before a reduction in density. However, axonal swellings can also be seen in normal individuals. Immunostaining for vasoactive intestinal polypeptide, substance P, or calcitonin gene-related proteins can be used to measure the density of sudomotor axons innervating sweat glands, piloerector nerves to hair follicles, and nerves to small arterioles. Myelin can be immunolabeled with antibodies directed against peripheral myelin protein 22 and myelin-associated glycoprotein.
Figure 3-35. Skin biopsy in small fiber neuropathy. A specimen obtained at the time of the patient’s first evaluation (A) shows a focal perivascular lymphocytic infiltrate (H&E, ×125). A section immunolabeled against protein gene product 9.5 reveals neural processes or axons (thick arrows) (B) showing an epidermal neurite with axonal swellings, which are abnormal (thin arrow). The density of nerve fibers is greater than normal (immunoperoxidase, ×500). A specimen obtained 11 months later (C) shows marked reduction in neurite density and axonal swelling (arrow) in a remaining neurite (×300). (Reproduced with permission from Drs. Thomas Smith and Lawrence Hayward, from Amato AA, Oaklander AL. Case 16–2004: A 76-year-old woman with numbness and pain in the feet and legs. N Engl J Med. 2004;350:2181–2189.)
SUMMARYMuscle, nerve, and skin biopsies for epidermal nerve fiber analysis can be useful in diagnosis of various neuromuscular conditions. The various histopathologic abnormalities that we mentioned are discussed in more detail in subsequent chapters with the diseases in which these appear. As with electrodiagnostic and other laboratory testing, these are only helpful in conjunction with a good clinical assessment and cogent differential diagnosis. Further, it is imperative that just as neuromuscular clinicians must be able to independently review and interpret results of electrodiagnostic testing, the same holds true for at least understanding biopsy reports. Whenever possible we would urge clinicians to review biopsy slides with their pathologists so that they can become more familiar with various disease processes and correlate the clinical and electrodiagnostic findings with the histopathology.
1. Banker BQ, Engel AG. Basic reactions of muscle. In: Engel AG, Franzini-Armstrong C, eds. Myology. 3rd ed. New York, NY: McGraw-Hill; 2004:691–747.
2. Carpenter S, Karpati G. Pathology of Skeletal Muscle. 2nd ed. New York, NY: Oxford; 2001.
3. De Girolami U, Frosch M, Amato AA. Biopsy of nerve and muscle. In: Samuels M, Feske S, eds. Office Practice of Neurology. 2nd ed. Philadelphia, PA: Harcourt Health Sciences;2003:217–225.
4. Dumitru D, Amato AA. Introduction to myopathies and muscle tissue’s reaction to injury. In: Dumitru D, Amato AA, Swartz MJ, eds. Electrodiagnostic Medicine. 2nd ed. Philadelphia, PA: Hanley & Belfus; 2002:1229–1264.
5. Engel AG. The muscle biopsy. In: Engel AG, Franzini-Armstrong C, eds. Myology. 3rd ed. New York, NY: McGraw-Hill; 2004:681–690.
6. Filosto M, Tonin P, Vattemi G, et al. The role of muscle biopsy in investigating isolated muscle pain. Neurology. 2007;68(3):181–186.
7. Cote AM, Jimenez L, Adelman LS, Munsat TL. Needle biopsy with the automatic Biopty instrument. Neurology. 1992;42:2212–2213.
8. Edwards RH, Round JM, Jones DA. Needle biopsy of skeletal muscle: A review of 10 years experience. Muscle Nerve. 1983:6(9):676–683.
9. Heckmatt JZ, Moosa A, Hutson C, Maunder-Sewry CA, Dubowitz V. Diagnostic needle muscle biopsy: A practical and reliable alternative to open biopsy. Arch Dis Child. 1994;59:528–532.
10. Magistris MR, Kohler A, Pizzolato G, et al. Needle muscle biopsy in the investigation of neuromuscular disorders. Muscle Nerve. 1988;21:194–200.
11. Colling-Saltin AS. Enzyme histochemistry on skeletal muscle of the human foetus. J Neurol Sci. 1978;39:169–185.
12. Engel AG, Banker BQ. Ultrastructural changes in diseased muscle. In: Engel AG, Franzini-Armstrong C, eds. Myology. 3rd ed. New York, NY: McGraw-Hill; 2004:749–887.
13. Johnson MA, Polgar J, Weightman D, Appleton D. Data on the distribution of fiber types in thirty-six human muscles. An autopsy study. J Neurol Sci. 1973;18:111–129.
14. Brooke MH, Engel WK. The histographic analysis of human muscle biopsies with regard to fiber types. 1. Adult male and female. Neurology. 1969;19:221–233.
15. Brooke MH, Engel WK. The histographic analysis of human muscle biopsies with regard to fiber types. 2. Diseases of the upper and lower motor neuron. Neurology. 1969;19:378–393.
16. Brooke MH, Engel WK. The histographic analysis of human muscle biopsies with regard to fiber types. 3. Myotonias, myasthenia gravis, and hypokalemic periodic paralysis. Neurology. 1969;19:469–477.
17. Brooke MH, Engel WK. The histographic analysis of human muscle biopsies with regard to fiber types. 4. Children’s biopsies. Neurology. 1969;19:591–605.
18. Metafora S, Felsani A, Cotrufo GF, et al. Neural control of gene expression in skeletal muscle fibers: The nature of the lesion in the muscular protein-synthesis machinery following denervation. Proc R Soc Lond B Biol Sci. 1980;209:239–255.
19. Bischoff R. Control of satellite cell proliferation. Adv Exp Med Biol. 1990;280:147–158.
20. Hilton DA, Jacob J, Househam L, Tengah C. Complications following sural and peroneal nerve biopsies. J Neurol Neurosurg Psychiatry. 2007;78:1271–1272.
21. Lacomis D. Clinical utility of peripheral nerve biopsy. Curr Neurol Neurosci Rep. 2005;5(1):41–47.
22. Ruth A, Schulmeyer FJ, Roesch M, Woertgen C, Brawanski A. Diagnostic and therapeutic value due to suspected diagnosis, longterm complications, and indication for sural nerve biopsy. Clin Neurol Neurosurg. 2005;107(3):214–217.
23. Schweikert K, Fuhr P, Probst A, Tolnay M, Renaud S, Steck AJ. Contribution of nerve biopsy to unclassified neuropathy. Eur Neurol. 2007;57(2):86–90.
24. Bouche P, Vallat JM. Neuropathies Périphériques. Paris: Doin; 1992.
25. Dyck PJ, Dyck PJB, Engelstad J. Pathologic alterations of nerves. In: Dyck PJ, Thomas PK, eds. Peripheral Neuropathy. 4th ed. Philadelphia, PA: WB Saunders; 2005:733–829.
26. Midroni G, Bilbao JM. Biopsy Diagnosis of Peripheral Neuropathy. Boston, MA: Butterworth-Heinemann; 1995.
27. Richardson EP Jr, De Girolami U. Pathology of the Peripheral Nerve. Philadelphia, PA: WB Saunders; 1995.
28. Ferreire G, Denef JF, Rodriguez J, Guzzetta F. Morphometric studies of normal sural nerves in children. Muscle Nerve 1985;8:697–704.
29. Amato AA, Oaklander AL. Case 16–2004: A 76-year-old woman with numbness and pain in the feet and legs. N Engl J Med. 2004;350:2181–2189.
30. Herrman DN, Griffin JW, Hauer P, Cornblath DR, McArthur JC. Intraepidermal nerve fiber density, sural nerve morphometry and electrodiagnosis in peripheral neuropathies. Neurology, 1999;53:1634–1640.
31. Holland NR, Stocks NR, Hauer P, Cornblath DR, Griffin JW, McArthur JC. Intraepidermal nerve fiber density in patients with painful sensory neuropathy. Neurology. 1997;48:708–711.
32. Holland NR, Crawford TO, Hauer P, Cornblath DR, Griffin JW, McArthur JC. Small-fiber sensory neuropathies: Clinical course and neuropathology of idiopathic cases. Ann Neurol. 1998; 44:47–59.
33. McArthur JC, Stocks EA, Hauer P, Cornblath DR, Griffin JW. Intraepidermal nerve fiber density: Normative reference range and diagnostic efficiency. Arch Neurol. 1998;55:1513–1520.
34. McCarthy BG, Hseih ST, Stocks A, et al. Cutaneous innervation in sensory neuropathies: valuation by skin biopsy. Neurology. 1995;45:1845–1855.
35. Periquet MI, Novak V, Collins MP, et al. Painful sensory neuropathy: Prospective evaluation of painful feet using electrodiagnosis and skin biopsy. Neurology. 1999;53:1641–1647.
36. Smith AG, Ramchandran P, Tripp S, Singleton JR. Epidermal nerve innervation in impaired glucose tolerance and diabetes-associated neuropathy. Neurology. 2001;57;1701–1704.
37. Sommer C, Lauria G. Skin biopsy in the management of peripheral neuropathy. Lancet Neurol. 2007;6:632–642.
38. Tobkin K, Guiliani MJ, Lacomis D. Comparison of different modalities for detection of small-fiber neuropathy. Clin Neurophys. 1999;110:1909–1912.
39. Walk D, Wendelschafer-Crabb G, Davey C, Kennedy WR. Concordance between epidermal nerve fiber density and sensory examination in patients with symptoms of idiopathic small fiber neuropathy. J Neurol Sci. 2007;255(1–2):23–26.
40. Wendelschafer-Crabb G, Kennedy WR, Walk D. Morphological features of nerves in skin biopsies. J Neurol Sci. 2006;242 (1–2):15–21.
41. England JD, Gronseth GS, Franklin G, et al. Evaluation of distal symmetric polyneuropathy: The role of autonomic testing, nerve biopsy, and skin biopsy (an evidence-based review). Muscle Nerve. 2009;39(1):106–115.
