Buckley, Anne F. MD, PhD; Bossen, Edward H. MD
The purpose of this review is to provide a framework for assessment of small blood vessels in the diagnostic evaluation of skeletal muscle biopsies. The requirement of skeletal muscle for effective oxygenation means that it is unusually vulnerable to ischemia or hypoxia. Vasculopathic conditions manifest readily in skeletal muscle for this reason, but it is often forgotten that muscle symptoms are frequently experienced by patients with systemic vasculopathies in the absence of a primary myopathy. Beyond the well-known finding of tubuloreticular inclusions, there are many other vascular features in skeletal muscle biopsies that can help narrow the differential diagnosis when interpreted in the context of fiber damage, stromal changes, and inflammatory cell patterns. Blood vessel changes may point the way to a correct diagnosis when other components of the biopsy are less revealing because of sampling error, preservation artifact, previous steroid therapy, or when the biopsy is taken early in the disease before other diagnostic features have developed. Evaluation of changes in skeletal muscle vasculature must also take into account factors common to many patients undergoing muscle biopsy, for example, diabetes mellitus, hypertension, and aging, so as not to overinterpret either the vascular findings or the background tissue changes caused by systemic conditions.
NORMAL SKELETAL MUSCLE MICROVASCULATURE
Primary arteries along the long axis of a muscle give rise to feeder arteries (1 or 2 per muscle) that send out primary, secondary, and tertiary branches; these last branches are small arterioles that run in septa between fascicles (1). Terminal arterioles branch off these and penetrate into fascicles; interconnecting capillary networks arising from the terminal arterioles surround individual muscle fibers. The capillary network of slow-twitch (Type 1) fibers, which have more mitochondria and oxidative enzymes, is up to 3-fold denser than that of fast-twitch (Type 2) fibers. Only a portion of the capillaries in normal skeletal muscle is functional at any given time, and arteriovenous shunts of precapillary vessels exist to reduce blood flow to some capillary networks. Postcapillary venules flow out to veins that run parallel to small arteries in septa.
Blood flow through muscle is controlled both by perfusion and vascular resistance and can vary 20- to 80-fold according to muscle requirements. Up to half of the vascular resistance is found in the feeder arteries. Vascular resistance varies inversely with the fourth power of the radius, so large changes in blood flow can be caused by small changes in vessel diameter. Skeletal muscle is richly innervated by the sympathetic nervous system, which affects vascular caliber via adrenergic receptors (1). Norepinephrine activates α receptors, which causes vasoconstriction, whereas epinephrine interacts with β2 receptors to cause vasodilation. Angiotensin II, produced by the renin-angiotensin system, constricts muscle arteries. Other vasoconstrictors include several varieties of endothelin. Vasodilators include nitric oxide, carbon monoxide, and prostacyclin. Nitric oxide is formed via the activity of endothelial nitric oxide synthase, which, in turn, is activated by vasodilators such as adenosine and acetylcholine. Hydrogen peroxide, generated by mitochondrial electron transfer, is another vasodilator (2).
Small septal arteries usually have 2 layers of smooth muscle cells (SMCs) (Fig. 1A). Smooth muscle cells are not simply contractile cells; they retain phenotypic plasticity and can change into other cell types, produce matrix material, and proliferate during vascular repair. They are also believed to be the source of multipotent stem cells (3, 4). Septal arteries also have elastic material in the basal lamina; this forms the internal elastic lamina, which is visible on toluidine blue–stained Epon sections as dense spots between the endothelial layer and the SMCs (5). The endothelial cells of septal arteries have high profiles and bulge into the lumen when the vessel is contracted.
Terminal arterioles have lumens of approximately 14 μm (Fig. 1B). These vessels are the last in the arterial hierarchy to have SMCs, which are present in a single layer. Terminal arterioles apparently function as precapillary sphincters (5), modulating blood flow to capillaries.
Normal capillaries have lumens of 5 to 10 μm and are lined by endothelial cells with flat profiles (Fig. 1C); the nuclei of the endothelial cells are oval or flattened in the natural curve of the lumen. Capillary diameters are larger in infants and children, and, in infants, endothelial cells are plumper and their nuclei are more evident (5). Exchange of fluids and proteins occurs at the capillary and postcapillary levels of microcirculation. Pores of 4 to 10 Å in endothelial cell membranes provide a pathway for oxygen, carbon dioxide, and very small water-soluble molecules, but most transvascular flow occurs through the tight junctions between endothelial cells. Tight junctions contain junctional protein complexes (composed of zona occludens, zonula adherens, and desmosomes) (6) and glycocalyx. The luminal surface of the endothelium is also lined by glycocalyx, which seems to serve as a mechanoreceptor and a restrictor of solutes (2).
The finding of internalized capillaries (i.e. capillaries within muscle fibers) is rare, and, when seen, it is more often associated with juvenile spinal muscular atrophy and Becker muscular dystrophy (7). These capillaries do not penetrate the sarcolemma of the fiber but run in an internalized extracellular space that seems to develop as a result of fiber splitting. They are more common in the gastrocnemius muscle and in males and are almost exclusive to Type I fibers. It is thought that the phenomenon of internalized capillaries is a result of hypoxia.
Postcapillary venules have lumens of 10 to 100 μm and have thinner walls than arterioles (1) (Fig. 1D). They also have pericytes and a discontinuous SMC layer. Septal veins are adjacent to septal arteries, have thinner walls and larger lumens than arteries, and are wrapped by SMCs and fibroblasts.
Pericytes are cells of mesodermal origin that are seen in all microvessel types. They have a variety of functions in angiogenesis, phagocytosis, and microcontractility. Pericytes wrap processes around capillaries (8) and actually bridge between them (9), reaching through the basal lamina to directly contact endothelial cells. Pericytes contain intermediate filaments, pinocytotic vesicles, and usually more glycogen than endothelial cells.
ULTRASTRUCTURAL FEATURES OF BLOOD VESSELS
Pinocytotic vesicles are approximately 50 to 75 nm in diameter and are visible only at the ultrastructural level (Fig. 2A) (10). They serve as a means of transport across the endothelial cell. The vesicles form from plasmalemmal invaginations on the luminal and abluminal surfaces and diffuse across the cytoplasm, fusing with the opposite plasmalemma and depositing solutes up to 150 Å. Other contents of endothelial cells include the Golgi apparatus, smooth and rough endoplasmic reticulum (ER), and glycogen granules.
Identification of endothelial cells is aided by the presence of Weibel-Palade bodies that store von Willebrand factor (vWF), a promoter of platelet aggregation (Fig. 2B). They are rod-shaped bodies approximately 0.1 × 3 μm and packed with tubular structures 150 Å thick that are aligned to the long axis of the rod (11). Their shape is determined by the molecular structure of vWF; thus, patients with vWF defects can have Weibel-Palade bodies with abnormal morphology (1).
Centrioles are physiologic structures that become evident during endothelial cell division (Fig. 2B, C). Their sizes vary from 0.5 to 2.0 μm according to their activity (12), and they are seen more frequently in pathologic states that cause proliferation of endothelial cells (13).
Intermediate filaments are a major feature of SMCs, which contain actin, myosin, and intermediate filaments of desmin and vimentin (14). Intermediate filaments are also present in pericytes. Endothelial cells also occasionally show filaments of approximately 10 nm (Fig. 2D) (5); this points to the plasticity of the cells that make up the walls of blood vessels and their ability to alter their phenotypes and function in regenerative states. The presence of filaments in an endothelial cell is, therefore, a nonspecific indicator of vascular regeneration.
Lipofuscin is often seen in skeletal muscle fibers as a nonspecific degenerative feature; it is part of normal aging and is composed of accumulated cellular by-products. It is seen less frequently in vascular endothelium (Fig. 2E), but here also it presumably indicates degenerative change.
Myelin figures are another nonspecific feature of degeneration frequently seen in skeletal muscle fibers (Fig. 2F). When seen in capillary endothelium, they are likewise a nonspecific indicator of vascular insult.
Paracrystalline inclusions (Fig. 2G) are structures of uncertain origin and composition that can be seen in conjunction with lipofuscin; a lysosomal origin has been proposed based on their ultrastructural morphology (15). They are occasionally seen in pediatric neuromuscular cases and rarely in adult patients. However, they have also been described in human fetuses (16), so they are neither specific nor necessarily pathologic.
CHANGES IN BASEMENT MEMBRANE
Blood vessel walls consist of endothelial cells, pericytes, and vascular SMCs, depending on the caliber of the vessel. These elements are bounded by a matrix, referred to as basement membrane (BM) or basal lamina, which serves as support and is involved in cell signaling. Most authors use the terms “BM” and “basal lamina” interchangeably, but some make a distinction between them. Those who make the distinction note that the term “BM” refers to the indistinct line outside of a cell membrane as seen by light microscopy. The term “basal lamina” includes the ultrastructural finding of a clear zone (lamina lucida) between the plasmalemma of a cell and the lamina densa (which is seen as BM by light microscopy); however, the lamina lucida is probably an artifact of fixation in standard electron microscopy (EM) because freeze-fracture EM does not demonstrate a structure corresponding to the lamina lucida (17–20). In some tissues, including skeletal muscle, the components of the BM include basal lamina plus a fibrillar reticular lamina (composed of Type III collagen) that lies outside the basal lamina (21). Following the most common usage (unless otherwise specified), in this review, we use the term “BM” to refer to the matrix around the cells of a blood vessel wall, understanding that, by EM, it would be referred to as “basal lamina” or “lamina densa.”
The BM is composed of laminin, collagen IV, perlecan, nidogens, and agrin, which are thought to be produced by endothelial cells (17). There are many types of laminin, which vary by body site; laminin-α4 and α5 chains predominate in blood vessels (22, 23). Laminin types also vary by stage of development; for example, β-1 chains are associated with vascular SMCs during development, but β-2 chains predominate in maturity (24). Pericytes and endothelial cells express fibronectin, collagen, and laminin receptors (25), which secure them to the BM.
The BM allows passage of water and small molecules. In normal adults, the capillary BM in skeletal muscle is approximately 50 μm thick. In the lower limbs of adults, the BM of capillaries is naturally thicker presumably because of higher venous pressures (5).
Thickening of the basal lamina occurs in normal aging. Although basal laminar thickening is a very common finding in arteries and capillaries, it is unusual to see it around larger septal veins. If basal laminar thickening around veins is seen in a muscle biopsy, the possibility of amyloid deposition should be considered (5). Basal laminar reduplication occurs in many myopathies and in neurogenic atrophies; it is thought to be caused by multiple cycles of capillary degeneration and regeneration (26).
Hypertensive patients frequently have thickened BMs (Fig. 3A); one cause of this is increased vascular resistance. Hypertension and obesity result in increased reactive oxygen species and may reduce production of nitric oxide (NO) by the endothelium, thereby increasing vascular resistance and impairing its vasodilator response (2). Another factor in hypertension is rarefaction of the microvasculature, which increases peripheral resistance. Essential (primary) hypertension is characterized by increased peripheral resistance in small arteries and arterioles. This normally occurs during aging, but if factors affecting angiogenesis are involved, it could begin early in life (27, 28). The increased wall-to-lumen ratio in essential hypertension is not caused by an increase in wall tissue but rather a rearrangement of tissue and inward wall remodeling that decreases lumen diameter (29). On the other hand, hypertrophic remodeling does occur in secondary hypertension.
Patients with diabetes mellitus, particularly Type 2, frequently have aging and hypertensive vascular changes, as well as additional damage caused by their diabetic condition. Hypertensive Type 2 diabetics demonstrate vascular SMC hyperplasia and hypertrophy (28). The vascular endothelial cell glycocalyx normally has abundant heparan sulfate proteoglycans, but these are reduced in diabetes, resulting in decreased NO production and increased macromolecular leakage through the endothelium (30). In diabetes mellitus, enhanced production of matrix proteins by transforming growth factor-β and inadequate matrix protein synthesis alter the thickness and composition of the BM (30); therefore, vascular BM thickening is common in patients with diabetes. Reduplication of the basal lamina is also frequently demonstrated in diabetes (Fig. 3B–D). The combined vascular changes in these patients are sometimes referred to as the “microangiopathy of diabetes.” An overlooked consequence of diabetic microangiopathy is skeletal muscle microinfarction, which needs to be distinguished from the microinfarction seen in dermatomyositis.
Peripheral arterial occlusive disease is also a cause of thickened basal lamina. Amputation specimens from patients with chronic peripheral arterial disease reveal a thickened (>1 μm) pericapillary BM coat, primarily composed of collagen IV (31). Ultrastructure studies of the coat in peripheral arterial disease demonstrate an amorphous matrix with occasional lamellae. This thickened coat correlates more with the history of peripheral arterial disease than with age or diabetes; cholesterol level and smoking do not correlate with the thickening.
At one time, the etiology of Duchenne muscular dystrophy (DMD) was thought possibly to be vascular (32). Vascular changes are seen in DMD, including thickened arterioles (33), and small foci of fiber degeneration at the periphery of muscle fascicles are common in early cases of DMD (34). The dystrophin-glycoprotein complex is involved in NO production, and disruption of this complex in DMD results in decreased NO and impaired vasodilation, causing repeated small areas of vascular injury (2, 35, 36). There has not been consistent evidence for a difference in blood flow between DMD patients and controls. Nonetheless, nonspecific capillary changes, for example, thinning of BM or replication of capillary BM, do occur in DMD (37, 38).
The skeletal muscle in patients with amyotrophic lateral sclerosis can demonstrate basal laminar thickening similar to that seen in diabetes mellitus (39). Capillary basal laminar thickening and reduplication can also be seen in multicore disease but not in central core myopathy (40). Thinning of basal lamina is seen in myotonic dystrophy (41). Reduplication of BM is seen often in dermatomyositis (DM) (42, 43). It is also seen in scleroderma and systemic lupus erythematosus (SLE) and has been reported in polymyositis (PM) (44). The BM area (as a percentage of capillary area) is increased in scleroderma and DM, but not in PM.
The elastic in the walls of septal arteries frequently shows changes in adults that suggest degeneration (Fig. 3E, F). In many cases, this material expands the basal lamina and can have the appearance of an immune deposit; however, we have seen this in so many muscle biopsies without being able to demonstrate an immune character that it seems to be a nonspecific effect of aging and/or common systemic vasculopathies such as hypertension. However, immune deposition should be considered because it can occur in this location.
Vascular deposition is a common finding in a number of neuromuscular disorders and can be detected by histochemical, immunohistochemical, immunofluorescent, or ultrastructural methods.
Direct immunofluorescence (DIF) is not often applied to skeletal muscle biopsies, but it can reveal vascular deposits in biopsies that do not show vascular abnormalities by light microscopy. In vasculitis, deposition of immunoglobulin and/or complement, especially IgG and C3, can be seen. Patients with collagen-vascular disorders show vascular staining patterns of granular deposits of immunoglobulin or complement (45). One meta-analysis indicated that vascular immunoglobulin and complement deposition are seen frequently in juvenile DM (JDM) (71%) and in 20% to 66% of other rheumatic disorders including adult DM (46). However, in PM, vascular deposition of immunoglobulin or complement is unusual (46–48). In DM, fibrinogen deposition can be seen in blood vessels as a manifestation of vascular damage (Fig. 4A). In JDM, the presence, in particular, of IgM, C3d, and fibrinogen in perimysial arteriolar vessels (as opposed to intrafascicular capillaries) correlates with a more aggressive course (49, 50). In addition, monoclonal gammopathies (paraproteinemias) can manifest in muscle biopsies; and DIF analysis of frozen muscle tissue can help identify and characterize them.
Vascular deposition of the complement membrane attack complex (MAC) can be detected using antibodies to C5b-9 (51). This is a common finding in DM, which is considered to be a complement-mediated microangiopathy that leads to capillary damage; clinical manifestations are foremost in skin and muscle but may also manifest in joints, lungs, heart, intestines, and brain. It is thought that capillary damage leads to hypoperfusion and inflammatory cell damage to muscle fibers. Support for a complement-mediated process through the classic pathway (i.e. initiated by antigen-antibody complexes) is provided by evidence that intravenous immunoglobulin works as an effective therapy in DM by preventing complement assembly and can reverse capillary loss and fiber atrophy (52). Membrane attack complex deposition in DM is visible before endothelial cell or muscle fiber damage or inflammation becomes evident, and it is seen in the walls of capillaries in minimally affected muscles; thus, it seems to represent one of the earliest events (43). Membrane attack complex deposition is seen more in areas with fiber necrosis and less in regenerating areas with atrophy (53). Membrane attack complex deposition in DM is seen on endomysial arterioles and capillaries and sometimes in larger perimysial arterioles, but not in venules. In JDM and early adult DM, it is seen throughout fascicles, with no relationship to myofiber necrosis. In established adult DM, MAC deposition is less consistent; it is seen more on capillaries and is associated with ischemic damage to fibers. An increased presence of MAC in DM presages a worse prognosis (54). Membrane attack complex deposits are not seen in PM (48). There is no MAC deposition in denervation even when vessel damage is present in those conditions, indicating that capillary damage per se does not lead to MAC deposition.
The presence of MAC on DIF is much more specific for DM than fibrinogen deposition and is the diagnostic marker of choice if available. However, in the absence of this marker, the more widely available (i.e. for renal biopsy evaluation) markers of vasculopathy can be useful.
Anti–signal recognition particle (anti-SRP) myopathy is a steroid-responsive, immune-mediated myopathy that demonstrates vascular MAC deposition and fiber necrosis, but with less inflammation (55). Another necrotizing pauci-immune myopathy responsive to steroids has been described that shows subendothelial deposition of MAC on BM elements rather than endothelial cells (56). Lastly, X-linked vacuolar myopathy has granular MAC complexes in perimysial blood vessels and endomysial capillaries, as well as in vacuoles in non-necrotic fibers (52), and can be considered if vascular and vacuolar MAC deposition is seen in a biopsy in the absence of muscle fiber necrosis.
Deposition of amyloid in skeletal muscle is rare, but it is important to identify it. Clinically, amyloidosis can manifest as pseudohypertrophy and weakness of muscle (57). It can be seen in patients with neoplasia, more often with hematologic malignancies such as multiple myeloma. Amyloid deposition can be perivascular or interstitial (58), and in blood vessels, it manifests as thickened basal lamina on hematoxylin and eosin (H&E) frozen sections and subendothelial or perivascular amyloid on EM (Fig. 4B). At high magnification, amyloid can be seen to consist of randomly arranged fibrils 50 to 250 Å (Fig. 4C, D) (13). It will often be accompanied by neurogenic atrophy caused by an amyloidogenic vascular neuropathy. Amyloid can be detected with Congo red staining under polarized light, but in practice, Congo red can have low sensitivity. Therefore, EM analysis should be considered if blood vessel walls seem thickened or if there is a clinical suspicion of amyloidosis.
Myopathies with pipe-stem capillaries can have varied histologic patterns, but the unifying feature is marked thickening of microvessel walls (venules, arterioles, and capillaries) that is prominent on H&E and periodic acid Schiff–stained sections (59–63). Other histologic features include endothelial cell swelling, microvascular deposition of MAC, and necrotic fibers with myophagocytosis. There is no perifascicular atrophy and little or no inflammation; no tubuloreticular inclusions (TRIs) are found on EM. The vascular thickening is caused by expansion of the basal lamina by amorphous material of unknown composition (Fig. 4E, F), and collagen has also been noted. There are no features of amyloid, and the appearance of the material is distinct from reduplication of basal lamina caused by repeated degeneration and regeneration of endothelial cells. The patients are not diabetic and do not have evident skin signs. Pipe-stem capillary myopathy is thought to represent an immune-mediated microangiopathy. It is not clearly a distinct entity, but it is a histopathologic pattern associated with neoplasia, autoimmune connective tissue disease, cerebral infarcts, and vasculitis (63). The associated myopathies respond to steroid therapy, so it is important to look for this feature in skeletal muscle biopsies.
VASCULAR ENDOTHELIAL CELL INCLUSIONS
Tubuloreticular inclusions consist of tubules approximately 250 Å in diameter, bundled in aggregates of 0.4 to 2.7 μm (42). They are found most frequently in vascular endothelial cells but can also be seen in pericytes and fibroblasts. Tubuloreticular inclusions can be identified in both intact and degenerating endothelial cells but are less likely to be visible in severely damaged capillaries. Endothelial cells with TRIs have fewer pinocytotic vesicles and can show increased microfilaments, which suggests that TRIs can be seen in both degenerating and regenerating endothelial cells. Tubuloreticular inclusions vary in appearance on EM (Fig. 5A–D): they can appear membrane bound, circumscribed by ER, or free in the cytoplasm. They can be contiguous with the Golgi apparatus and nuclear membrane but are also seen close to cell junctions.
It is still not known whether TRIs are defective viral structures or abnormally reacting organelles. It seems clear that interferon signaling can induce TRIs because treatment of cultured lymphocytes with interferon-α (a therapy for hepatitis B) results in their appearance (64), and molecules upregulated by interferons are expressed in both endothelial cells and muscle fibers in DM (65).
Tubuloreticular inclusions are considered a reliable way to discriminate between DM and other inflammatory myopathies on skeletal muscle biopsies. They are seen consistently in the microvasculature of both juvenile and adult DM (42, 44, 66, 67), and their presence in an immune-mediated inflammatory myopathy is considered diagnostic of DM in the appropriate clinical context. However, TRIs have been seen in capillary endothelial cells of skeletal muscle from HIV-infected patients who developed myopathies on azidothymidine therapy (68). Tubuloreticular inclusions are frequently seen in SLE and can also be seen in patients with other connective tissue diseases, including scleroderma and Sjögren syndrome, in the absence of any skin findings of DM (69).
It should be borne in mind that TRIs have been identified in rare cases with clinical and EM features of inclusion-body myositis (IBM), including the diagnostic inclusion-body filaments (70). They have also been described in PM (69), although the question of definition arises in these cases, as diagnoses of PM are often given to patients who lack the characteristic clinical features of DM.
Other vascular endothelial inclusions that have been reported in similar contexts include cylindrical confronting cisternae (71) or striated membranous structures (72). These structures are considered to have a similar diagnostic significance to TRIs.
Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is a systemic angiopathy caused by mutations in the NOTCH3 gene that lead to degeneration of SMCs and pericytes in small arteries and arterioles (73, 74). It is a rare disease, but it can be diagnosed on muscle biopsy using EM. The diagnostic feature of CADASIL is the presence of granular osmiophilic material in the basal lamina around SMCs and pericytes in blood vessels (75, 76).
ENDOTHELIAL CELL REACTION, HYPERTROPHY, HYPERPLASIA, DEGENERATION, AND REGENERATION
Endothelial cell reaction is a subtle nonspecific change in which endothelial cells extend cytoplasmic extensions into the vascular lumen (Fig. 6A), indicating increased adhesive and migration interactions with blood cells. It is a common feature in inflammatory conditions. In DM, the earliest microscopic findings are of capillary endothelial cell reaction and hypertrophy, in which endothelial cells swell to a plump or cuboidal morphology and narrow the lumen of capillaries, veins, and small arteries (Fig. 6B) (42). The nuclei of these endothelial cells also change from oval and uniform without nucleoli to large pleomorphic nuclei with large nucleoli. Pericytes also show hypertrophic changes. Hypertrophy may also indicate endothelial cell regeneration, with increased presence of smooth and rough ER, Golgi apparatus, and mitochondria, with fewer pinocytosis vesicles than normal. In JDM, the gap between endothelial cells in intramuscular capillaries increases from 100 to 200 to 300 to 800 Å, and the tight junctions are deficient or absent. These changes are usually present in DM even when TRIs and fiber damage are not evident and are seen in endomysial, perimysial, and epimysial vessels. One study of clinically defined cases of DM, SLE, PM, and scleroderma found that swollen endothelial cells and pericytes could be seen in all of the entities but with varied frequency (DM, 20%; PM, 10%; scleroderma/SLE, 4%) (44). Anti-SRP myopathy also shows hypertrophy of capillary endothelial cells (55).
Swelling of capillary endothelial cells is also seen in mitochondrial myopathies, but not core myopathies (77). Endothelial cell swelling has been noted in established DMD but is not an early finding and is not generally noted in muscular dystrophies.
Endothelial cell degeneration (Fig. 6C) must be distinguished from preservation artifact by considering the relative preservation of adjacent structures. Microvacuoles are an early sign of in vivo endothelial cell degeneration and are, perhaps, caused by altered permeability of the endothelial plasma membrane (43). Dermatomyositis frequently demonstrates widespread vascular endothelial cell degeneration, and autophagy can also be seen (42). Degenerating endothelial cells lack organelles, except for a few polyribosomes, and junctions are lost. Platelet thrombi can be found in association with damaged endothelial cells, and erythrocyte diapedesis can be seen. It should be noted that denervation atrophy can also show necrotic endothelial cells. In rapidly progressive denervation in which fiber type grouping is not yet established, the extent of capillary damage can resemble that seen in DM; accurate clinical history is a requirement.
“Hyperplasia” refers to an increase in the number of endothelial cells in enlarged capillary walls. In normal capillaries, there is an average of 1.6 tight junctions per cross section; this has been found to increase to 2.6 in scleroderma and DM but is not increased in PM or SLE (44). Mean capillary area (luminal + endothelial) is markedly larger in scleroderma and DM; it is also increased in SLE, but not in PM.
Although nonspecific, the endothelial features discussed in this section are useful indicators of vascular insult, and when considered in the context of other histologic features and clinical history, they can narrow the differential diagnosis. In a patient with an immune-mediated inflammatory myopathy, the presence of extensive capillary degeneration and hyperplasia supports a capillaritis like DM over a diagnosis of PM or IBM, even if TRIs are not visible.
CAPILLARY LOSS AND GAIN
Normally, each skeletal muscle fiber is adjacent to 2 (Type 2 fibers) to 4 (Type 1 fibers) capillaries in a transverse section of muscle (78); an example of normal capillary presence is seen in Figure 7A. The number of capillaries per fiber correlates to the number of mitochondria, and these are more abundant in Type 1 fibers. Endurance training increases the number of capillaries per fiber (79). Capillary density refers to the number of capillaries per square millimeter of muscle fiber area; this density increases when muscle fibers atrophy or drop out, unless capillaries are also lost. Capillary loss (or capillary necrosis) is indicated on tissue sections by loss of endothelial CD31 immunostaining (Fig. 7B) or lectin binding and is confirmed by the finding of empty loops of BM in the endomysium on EM (Fig. 7C). Extravasated red blood cells are sometimes associated with BM remnants.
An apparent reduction in capillary density is common in DM, and the remaining capillaries are enlarged (Fig. 7B) (54, 80). Capillary loss is also noted in anti-SRP, scleroderma, and SLE (55). Capillary loss and necrosis are not usually described in PM (54, 81). In 1 study of different categories of neuromuscular disease that considered the linear relationship between capillary number per fiber and fiber size, DM and PM had decreased capillaries per fiber, but the loss was more pronounced in DM (82). Capillary loss is more evident in DM relative to PM because, in DM, the loss of capillaries is much more extensive than the loss of muscle fibers. Capillary loss in DM is associated with more severe fiber damage (80) and is seen in areas distant from intermediate-sized perimysial arteries, suggesting a watershed effect of ischemia. However, this capillary loss is not consistently related to perifascicular atrophy (54); therefore, the assumption that perifascicular atrophy is caused by ischemia has been challenged. Perifascicular atrophy is not seen in diabetic microinfarction or vasculitis (83), and animal models of ischemia do not show perifascicular atrophy (84). One study of microarterial embolization in rabbits with 20 to 80 μm dextran particles showed fiber necrosis in the center of fascicles rather than the periphery (33). There is evidence to support the idea that 2 separate processes are causing the muscle changes in DM: (i) zonal ischemia and fiber necrosis caused by microvascular damage; and (ii) perifascicular atrophy mediated by secretion of soluble factors, including interferon signaling from plasmacytoid dendritic cells in perifascicular areas (85, 86).
Duchenne muscular dystrophy has been shown to have reduced capillaries per fiber (82), which may be caused by interstitial fibrosis, as connective tissue is thought to inhibit angiogenesis. Capillary necrosis is not seen in DMD, but the presence of reduplicated basal lamina in DMD does suggest endothelial degeneration and regeneration in this disease. In contrast, limb girdle dystrophies and IBM have increased numbers of capillaries per fiber.
Capillary necrosis also occurs in denervation atrophy (78). In established neurogenic atrophy, there is an overall increase in capillary density, even though capillary loss occurs, because of shrinkage of muscle fibers and decreased fascicle area. In patients with lower motor neuron disease, there seems to be a reactive hyperperfusion of denervated muscle, so this loss of capillaries may be a compensatory mechanism; however, this mechanism is not understood. Studies in rats show that the loss of capillaries in denervation occurs much more rapidly than the fiber atrophy and that Type 2 fibers are affected first because they are less well vascularized than Type 1 fibers (87); this can result in selective Type 2 fiber atrophy in early neurogenic atrophy.
Ragged red fibers have higher numbers of capillaries around them; this may be caused by increased capillary growth secondary to impaired muscle oxidative phosphorylation (88). In patients with systemic mitochondrial disorders, increased capillary numbers per fiber can be seen throughout the biopsy by CD31 immunostaining (Fig. 7D); presumably, this is via a similar mechanism.
Interpretation of capillary density and capillary numbers per fiber must be done with caution and with due regard to the relative fiber sizes and fiber type ratios in the section being examined. Published studies of these parameters have involved precise quantification, which is not realistic for routine diagnostic use. Nonetheless, in the context of other features in the biopsy, consideration of capillary density changes and numbers of capillaries per fiber may may narrow the differential diagnosis or suggest diagnostic possibilities in biopsies that show no other findings by light microcopy. For routine application, we consider that if multiple fibers in a field have 5 or more adjacent capillaries or if multiple fibers in a field have 1 or no adjacent capillaries, an abnormal capillary distribution should be considered in determining the differential diagnosis and directing the histologic workup. This determination can be made regardless of fiber typing but will be more valuable and precise if fiber ratio is also borne in mind.
ALKALINE PHOSPHATASE STAINING
Alkaline phosphatase staining in frozen sections of skeletal muscle is a very useful indicator of both vascular and muscle damage. Alkaline phosphatase activity is a marker for cells with active cellular transport and is normally found in the cytoplasm of neutrophils, osteoblasts, vascular endothelial cells, and some lymphocytes (89). In normal skeletal muscle, alkaline phosphatase activity is found in the plasma membranes of cells in which active transport processes occur, such as the endothelium of arterioles and arterial capillaries, in their ER, Golgi apparatus, and pinocytotic vesicles. It is also present in the pericytes of human muscle (90) and skeletal muscle progenitors (91). Alkaline phosphatase activity is absent in human venous endothelium; this is true of most species used for research, with the exception of birds (92). On frozen sections of normal skeletal muscle, alkaline phosphatase activity is not visible in capillaries; the histologic stain highlights only arterioles. In pathologic states, such as in active inflammatory myopathies, however, the alkaline phosphatase enzyme may leak out from capillaries into the interstitium because of increased vascular permeability or endothelial cell damage or its expression may increase in the stromal cells of the perimysium. Perimysial areas may therefore be stained in inflammatory myopathies. In particular, alkaline phosphatase staining in a perifascicular pattern (Fig. 8A, B) is a feature of paraneoplastic myopathies (93), as well as in DM, which has a strong association with neoplasia. A perifascicular pattern of alkaline phosphatase in a muscle biopsy therefore indicates possible neoplasia, and the treating physician should specifically be alerted to this.
Another abnormal pattern of alkaline phosphatase staining emerges when enzymatic activity becomes evident in capillaries on histologic staining (Fig. 8C). This pattern is less specific and can be seen in many inflammatory myopathies and metabolic disorders, as well as in patients with systemic vasculopathies. However, in a muscle biopsy with few changes on light microscopy, the expression of alkaline phosphatase in capillaries is a useful indicator to take a closer look with other stains and by EM.
Perivascular mononuclear inflammation is often seen in inflammatory myopathies and sometimes results in a non-necrotizing secondary vasculitis (Fig. 9A). This feature is more evident in DM than other inflammatory myopathies but is not reliable in making the distinction because perivascular inflammation can be seen in any muscle biopsy in which tissue damage has occurred. However, it is useful to do immunostaining to subtype the perivascular lymphocytes: a predominance of CD20-positive B cells in a perivascular inflammatory population (Fig. 9A inset) is supportive of a diagnosis of DM over PM or IBM because it indicates a humorally mediated vasculopathic myopathy, as opposed to a cytotoxic inflammatory myopathy.
Necrotizing vasculitis, on the other hand, demonstrates fibrinoid necrosis, transmural infiltration by inflammatory cells, and endothelial layer disruption (83) (Fig. 9B). Primary vasculitides that manifest in muscle include microscopic polyarteritis, polyarteritis nodosa, Churg-Strauss syndrome, and Wegener granulomatosis. In a workup for vasculitis, sural nerve biopsies are done more frequently, but sometimes an accompanying gastrocnemius muscle biopsy will show vasculitic neuropathy and/or vasculitis when the nerve biopsy is unrevealing; therefore, there is an increased diagnostic yield if both are done.
Stroke-like episodes can be associated with mitochondrial disorders, most frequently in the disorder known as mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS). These episodes are thought to result from a poorly understood mitochondrial angiopathy, perhaps leading to vasogenic edema (94). Abnormal mitochondria have been found in the brains and in the muscle fibers and endomysial capillary endothelial cells of MELAS patients (95). A recent study in infants with mitochondrial disorders showed abnormal mitochondria frequently in the endothelium of intramuscular capillaries (96). These endothelial cells showed reactive changes, including extension of fingers of cytoplasm into the lumen. Mitochondria in the vascular endothelium more frequently showed abnormalities than mitochondria in the muscle fibers (50%–70% vs 5%–10%), and those mitochondrial abnormalities were more pronounced in endothelial cells. Increased staining for succinic acid dehydrogenase can be seen in perimysial arteries and arterioles in mitochondrial disorders because of a reactive increase in mitochondria that occurs predominantly in SMCs (97). In addition, some lysosomal storage diseases (e.g. fucosidosis, Fabry disease, Sandhoff disease, and Batten disease) can show characteristic deposits in endothelial cells of capillaries and larger vessels (5).
TECHNICAL CONSIDERATIONS IN EVALUATING SKELETAL MUSCLE VASCULATURE
Optimal evaluation of skeletal muscle for vascular pathology requires both frozen and glutaraldehyde-fixed tissue. Freezing preserves the enzymatic activity required for alkaline phosphatase and succinic acid dehydrogenase staining. Frozen tissue is also ideal for DIF analysis of vascular deposits. It is always worthwhile to set aside some frozen muscle tissue in case it is needed for further workup later. Even if not needed for enzyme histochemical analysis, skeletal muscle biopsy tissue is increasingly used for follow-up biochemical studies; therefore, having stored frozen tissue available can be of immense value to the patient. With this in mind, embedding of frozen muscle tissue should be done in gum tragacanth rather than synthetic mounting media because this gum is harmless to mitochondrial enzymes and other clinically relevant biochemical activities.
We are aware that electron microscopic analysis is not used as often as in the past, but it is of great value in the evaluation of vasculature. Even if full EM is not done, blood vessels can be quite effectively assessed on plastic-embedded thick (semithin) sections because the light microscopic resolution of plastic sections is much higher than that of frozen or paraffin-embedded tissue, and preservation is often better. For instance, capillary degeneration or loss is often apparent on plastic-embedded sections, and perivascular deposits are readily identified. Even if EM is not done or is not available, it is worthwhile to put a small piece of muscle tissue into glutaraldehyde for possible future application or outside consultation on the case.
Paraffin-embedded tissue can be used to detect vasculitis, perivascular inflammation, and perivascular deposits, but it is no better for these purposes than a combination of frozen and glutaraldehyde-fixed tissue. For rapid preliminary analysis of skeletal muscle biopsies for inflammation, infarction, or neoplasia, paraffin sections can be replaced by frozen H&E-stained sections. An important disadvantage of paraffin embedding is that muscle tissue is thereby lost to future biochemical analysis; therefore, we do not use this embedding method for skeletal muscle biopsies.
Blood vessels, like all biologic structures, have a limited repertoire of histologically recognizable reactions to insult. The pathologic features discussed in this review will be useful only when interpreted in the context of accurate clinical history and a careful examination of other histologic features of the muscle biopsy. As part of the analysis of skeletal muscle biopsies, blood vessel findings can provide useful information that allows pathologists examining muscle biopsies to narrow their differential diagnoses and provide more accurate guidance to treating physicians.
1. Cheville NF. Ultrastructural Pathology: The Comparative Cellular Basis of Disease. 2nd ed. Ames, IA: Wiley-Blackwell, 2009
2. Korthius RJ. Skeletal Muscle Circulation. San Rafael, CA: Morgan & Claypool Life Sciences, 2011
3. Alexander MR, Owens GK. Epigenetic control of smooth muscle cell differentiation and phenotypic switching in vascular development and disease. Annu Rev Phys 2012; 74: 13–40
4. Crisan M, Yap S, Casteilla L, et al. A perivascular origin for mesenchymal stem cells in multiple human organs. Cell Stem Cell 2008; 3: 301–13
5. Carpenter S, Karpati G. Pathology of Skeletal Muscle. 2nd ed. Oxford, UK: Oxford University Press, 2001
6. Gumbiner B. Structure, biochemistry, and assembly of epithelial tight junctions. Am J Physiol 1987; 22: C749–58
7. Wolf R, Goebel HH, Gutmann L, et al. Capillaries within human skeletal muscle fibers. Pathol Res Pract 1991; 187: 857–63
8. Sims DE. The pericyte—A review. Tissue Cell. 1986; 18: 153–74
9. Williamson JR, Tilton RG, Kilo C, et al. Immunofluorescent imaging of capillaries and pericytes in human skeletal muscle and retina. Microvasc Res 1980; 20: 233–41
10. Moore DH, Ruska H. The fine structure of capillaries and small arteries. J Biophys Biochem 1957; 3: 457–75
11. Weibel ER, Palade GE. New cytoplasmic components in arterial endothelia. J Cell Biol 1964; 23: 101–12
12. Bystrevskaya VB, Lichkun VV, Krushinsky AV, et al. Centriole modification in human aortic endothelial cells. Struct Biol 1992; 109: 1–12
13. Mair WGP, Tome FMS. Atlas of the Ultrastructure of Diseased Human Muscle
. Edinburgh, UK: Churchill Livingstone, 1972
14. Kierszenbaum AL, Tres LL. Histology and Cell Biology: An Introduction to Pathology. 3rd ed. Philadelphia, PA: Elsevier, 2012
15. Sengel A, Stoebner P. Pseudocrystalline inclusions in muscular endothelial cells. Virchows Archiv B Cell Pathol 1972; 10: 354–58
16. Spear GS, Slusser RJ, Garvin AJ, et al. A cytoplasmic body in human fetal endothelium. Am J Pathol 1975; 78: 333–37
17. Hohenester E, Yurchenco PD. Laminins in basement membrane assembly. Cell Adhesion Migration 2013; 7: 56–63
18. Yurchenco PD, Patton BL. Developmental and pathogenic mechanisms of basement membrane assembly. Curr Pharm Des 2009; 15: 1277–94
19. Miosge N. The ultrastructural composition of basement membranes in vivo. Histol Histopathol 2001; 16: 1239–48
20. Chan FL, Inoue S, Leblond CP. Cryofixation of basement membranes followed by freeze substitution or freeze drying demonstrates that they are composed of a tridimensional network of irregular cords. Anat Rec 1993; 235: 191–205
21. Sanes JR. The basement membrane/basal lamina of skeletal muscle. J Biol Chem 2003; 278: 12601–4
22. Yousif LF, DiRusso J, Sorokin L. Laminin isoforms in endothelial and perivascular basement membranes. Cell Adh Migr 2013; 7: 101–10
23. Simon-Assmann P, Orend G, Mammadova-Bach E, et al. Role of laminins in physiological and pathological angiogenesis. Int J Dev Biol 2012; 55: 455–65
24. Glukhova M, Kotelianshy V, Fondacci C, et al. Laminin variants and integrin laminin receptors in developing and adult human smooth muscle. Dev Biol 1993; 157: 437–47
25. Silva R, D’Amico G, Hodivala-Dilke KM, et al. Integrins: The key to unlocking angiogenesis. Arterioscler Thromb Vasc Biol 2008; 28: 1703–13
26. Vracko R. A comparison of the microvascular lesions in diabetes mellitus with those in normal aging. J Am Geriatr Soc 1982; 30: 201–5
27. Humar R, Zimmerli L, Battegay E. Angiogenesis and hypertension: An update. J Human Hypertension 2009; 23: 773–82
28. Rizzoni D, Rosei EA. Small artery remodeling in diabetes mellitus. Nutr Metab Cardiovasc Dis 2009; 19: 587–92
29. Mulvaney MJ. Small artery remodeling in hypertension. Basic Clin Pharmacol Toxicol 2012; 110: 49–55
30. Bakker W, Eringa EC, Sipkema P, et al. Endothelial dysfunction and diabetes: Roles of hyperglycemia, impaired insulin signaling and obesity. Cell Tissue Res 2009; 335: 165–89
31. Baum O, Djonov V, Ganster M, et al. Arteriolization of capillaries and FGF-2 up-regulation in skeletal muscles of patients with chronic peripheral arterial disease. Microcirculation 2005; 12: 527–37
32. Rowland LP. Pathogenesis of muscular dystrophies. Arch Neurol 1976; 33: 315–21
33. Hathaway PW, Engel WK, Zellweger H. Experimental myopathy after microarterial embolization: Comparison with childhood X-linked pseudohypertrophic muscular dystrophy. Arch Neurol 1970; 22: 365–78
34. Engel WK. Muscle biopsies in neuromuscular diseases. Pediatr Clin North Am 1967; 14: 963–95
35. Ramachandran J, Schneider JS, Crassous PA, et al. Nitric oxide signaling pathway in Duchenne muscular dystrophy mice: Up-regulation of L-arginine transporters. Biochem J 2013; 449: 133–42
36. Brenman JE, Chao DS, Xia H, et al. Nitric oxide synthase complexed with dystrophin and absent from skeletal muscle sarcolemma in Duchenne muscular dystrophy. Cell 1995; 82: 743–52
37. Leinonen H, Juntunen J, Somer H, et al. Capillary circulation and morphology in Duchenne muscular dystrophy. Eur Neurol 1979; 18: 249–55
38. Musch BC, Papapetropoulos TA, McQueen DA, et al. A comparison of the structure of small blood vessels in normal, denervated and dystrophic human muscle. J Neurol Sci 1975; 26: 221–34
39. Mueller PS, Quick DT. Studies of glucose, insulin, and lipid metabolism in amyotrophic lateral sclerosis and other neuromuscular disorders. J Lab Clin Med 1970; 76: 190–201
40. Kobayashi Y, Suzuki H, Iinuma K, et al. Endothelial alterations of skeletal muscle capillaries in childhood myopathies. J Exp Med 1983; 140: 381–89
41. Olson ND, Nuttall FQ, Akhouri S, et al. Thin muscle capillary basement membranes in myotonic dystrophy. Diabetes 1979; 28: 686–89
42. Banker BQ. Dermatomyositis of childhood, ultrastructural alterations of muscle and intramuscular blood vessels. J Neuropathol Exp Neurol 1975; 34: 46–75
43. De Visser M, Emslie-Smith AM, Engel AG. Early ultrastructural alterations in adult dermatomyositis. Capillary abnormalities precede other structural changes in muscle. J Neurol Sci 1989; 94: 181–92
44. Jerusalem F, Rakusa M, Engel AG, et al. Morphometric analysis of skeletal muscle capillary ultrastructure in inflammatory myopathies. J Neurol Sci 1974; 23: 392–402
45. Oxenhandler R, Adelstein EH, Hart MN. Immunopathology of skeletal muscle. The value of direct immunofluorescence in the diagnosis of connective tissue disease. Hum Pathol 1977; 8: 321–28
46. Oxenhandler R, Hart MN. Skeletal muscle immunopathology. Hum Pathol 1983; 14: 326–37
47. Heffner RR Jr, Barron SA, Jenis EH, et al. Skeletal muscle in polymyositis: Immunohistochemical study. Arch Pathol Lab Med 1979; 103: 310–13
48. Whitaker JN, Engel WK. Vascular deposits of immunoglobulin and complement in idiopathic inflammatory myopathy. N Engl J Med 1972; 286: 333–38
49. Crow WE, Bove KE, Levinson JE, et al. Clinical and pathogenetic implications of histopathology in childhood polydermatomyositis. Arthritis Rheum 1982; 25: 126–39
50. Wargula JC, Lovell DJ, Passo MH, et al. What more can we learn from muscle histopathology in children with dermatomyositis/polymyositis? Clin Exp Rheumatol 2006; 24: 333–43
51. Jain A, Sharma MC, Sarkar C, et al. Detection of the membrane attack complex as a diagnostic tool in dermatomyositis. Acta Neurol Scand 2011; 123: 122–29
52. Mendell JR, Garcha TS, Kissel JT. The immunopathogenic role of complement in human muscle disease. Curr Opin Neurol 1996; 9: 226–34
53. Kissel JT, Mendell JR, Rammohan KW. Microvascular deposition of complement membrane attack complex in dermatomyositis. N Engl J Med 1986; 314: 329–34
54. Emslie-Smith AM, Engel AG. Microvascular changes in early and advanced dermatomyositis: A quantitative study. Ann Neurol 1990; 27: 343–56
55. Miller T, Al-Lozi MT, Lopate G, et al. Myopathy with antibodies to the signal recognition particle: Clinical and pathological features. J Neurol Neurosurg Psych 2002; 73: 420–28
56. De Bleecker J, Vervaet V, Van den Bergh P. Necrotizing myopathy with microvascular deposition of the complement membrane attack complex. Clin Neuropathol 2004; 23: 76–79
57. Jeneekens FG, Wokke JH. Proximal weakness of the extremities as main feature of amyloid myopathy. J Neurol Neurosurg Psychiatry 1987; 50: 1353–58
58. Prayson RA. Amyloid myopathy: Clinicopathologic study of 16 cases. Hum Pathol 1998; 29: 463–68
59. Emslie-Smith AM, Engel AG. Necrotizing myopathy with pipestem capillaries, microvascular deposition of the complement membrane attack complex (MAC), and minimal cellular infiltration. Neurology 1991; 41: 936–39
60. Authier FJ, Kondo H, Ghnassia RT, et al. Necrotizing myopathy with pipestem capillaries and minimal cellular infiltration: A case associated with cutaneous signs of dermatomyositis. Neurology 1996; 46: 1448–51
61. Reimann J, Kornblum C, Tolksdorf K, et al. Myopathy and neuropathy with pipestem capillaries and vascular activated complement deposition. Neurology 2011; 77: 401–3
62. Janzen RW, Weber M, Schlote W. Non-necrotizing autoimmune myopathy with pipestem type capillaries—A light and electron microscopic study. Clin Neuropathol 2012; 31: 81–86
63. Schröder NW, Goebel HH, Brandis A, et al. Pipestem capillaries in necrotizing myopathy revisited. Neuromuscul Disord 2013; 23: 66–74
64. Grimley PM, Davis GL, Kang YH, et al. Tubuloreticular inclusions in peripheral blood mononuclear cells related to systemic therapy with alpha-interferon. Lab Invest 1985; 52: 638–49
65. Greenberg SA. Dermatomyositis and Type 1 interferons. Curr Rheumatol Rep 2010; 12: 198–203
66. Norton WL, Velayos E, Robison L. Endothelial inclusions in dermatomyositis. Ann Rheum Dis 1970; 29: 67–72
67. Carpenter S, Karpati G, Rothman S, et al. The childhood type of dermatomyositis. Neurology 1976; 26: 952–62
68. Lane RJ, McLean KA, Moss J, et al. Myopathy in HIV infection: The role of zidovudine and the significance of tubuloreticular inclusions. Neuropathol Appl Neurobiol 1993; 19: 406–13
69. Bronner IM, Hoogendijk JE, Veldman H, et al. Tubuloreticular structures in different types of myositis: Implications for pathogenesis. Ultrastruct Path 2008; 32: 123–26
70. Katzberg HD, Munoz DG. Tubuloreticular inclusions in inclusion body myositis. Clin Neuropathol 2010; 29: 262–66
71. Bockus D, Remington F, Luu JY, et al. Induction of cylindrical confronting cisternae (AIDS inclusions) in Daudi lymphoblastoid cells by recombinant alpha-interferon. Human Pathol 1988; 19: 78–82
72. Fidzianska A, Goebel HH. Tubuloreticular structures (TRS) and cylindric confronting cisternae (CCC) in childhood dermatomyositis. Acta Neuropathol 1989; 79: 310–16
73. Arboleda-Velasquez JF, Manent J, Lee JH, et al. Hypomorphic Notch 3 alleles link Notch signaling to ischemic cerebral small-vessel disease. Proc Natl Acad Sci USA 2011; 108: E128–35
74. Dziewulska D, Lewandowska E. Pericytes as a new target for pathological processes in CADASIL. Neuropathology 2012; 32: 515–21
75. Mayer M, Straube A, Bruening R, et al. Muscle and skin biopsies are a sensitive diagnostic tool in the diagnosis of CADASIL. J Neurol 1999; 246: 526–32
76. Lewandowska E, Dziewulska D, Parys M, et al. Ultrastructure of granular osmiophilic material deposits (GOM) in arterioles of CADASIL patients. Folia Neuropathol 2011; 49: 174–80
77. Kobayashi Y, Suzuki H, Iinuma K, et al. Endothelial alterations of skeletal muscle capillaries in childhood myopathies. Tohuku J Exp Med 1983; 140: 381–89
78. Carpenter S, Karpati G. Necrosis of capillaries in denervation atrophy of human skeletal muscle. Muscle Nerve 1982; 5: 250–54
79. Ingjer F. Effects of endurance training on muscle fibre ATPase activity, capillary supply and mitochondrial content in man. J Physiol 1979; 294: 419–32
80. Pestronk A, Schmidt RE, Choksi R. Vascular pathology in dermatomyositis and anatomic relations to myopathology. Muscle Nerve 2010; 42: 53–61
81. Carpenter S, Karpati G. The major inflammatory myopathies of unknown cause. Pathol Annu 1981; 16: 205–37
82. Carry MR, Ringel SP, Starcevich JM. Distribution of capillaries in normal and diseased human skeletal muscle. Muscle Nerve 1986; 9: 445–54
83. Prayson RA. Skeletal muscle vasculitis exclusive of inflammatory myopathic conditions: A clinicopathologic study of 40 patients. Hum Pathol 2002; 33: 989–95
84. Karpati G, Carpenter S, Melmed C, et al. Experimental ischemic myopathy. J Neurol Sci 1974; 23: 129–61
85. Greenberg SA, Pinkus JL, Pinkus GS, et al. Interferon-alpha/beta-mediated innate immune mechanisms in dermatomyositis. Ann Neurol 2005; 57: 664–78
86. Greenberg SA. Proposed immunologic models of the inflammatory myopathies and potential therapeutic implications. Neurology 2007; 69: 2008–19
87. Borisov AB, Huang SK, Carlson BM. Remodeling of the vascular bed and progressive loss of capillaries in denervated skeletal muscle. Anat Rec 2000; 258: 292–304
88. Taivassalo T, Ayyad K, Haller RG. Increased capillaries in mitochondrial myopathy: Implications for the regulation of oxygen delivery. Brain 2012; 135: 53–61
89. Greer P, Foerster J, Rodgers GM, et al. eds. Wintrobe’s Clinical Hematology. 12th ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2009
90. dos Santos Goldenberg RC, de Carvalho ACC, eds. Resident Stem Cells and Regenerative Therapy. Oxford, UK: Academic Press (Elsevier), 2012
91. Abraham D, ed. Translational Vascular Medicine: Pathogenesis, Diagnosis, and Treatment. London, UK: Springer-Verlag, 2012
92. Grim M, Carlson BM. Alkaline phosphatase and dipeptidylpeptidase IV staining of tissue components of skeletal muscle: A comparative study. J Histochem Cytochem 1990; 38: 1907–12
93. Levin MI, Mozaffar T, Al-Lozi MT, et al. Paraneoplastic necrotizing myopathy: Clinical and pathological features. Neurology 1998; 50: 764–67
94. Finsterer J. Stroke and stroke-like episodes in muscle disease. Open Neurol J 2012; 6: 26–36
95. Fujii T, Okuno T, Ito M, et al. MELAS of infantile onset: Mitochondrial angiopathy or cytopathy? J Neurol Sci 1991; 103: 37–41
96. Sarnat HB, Flores-Sarnat L, Casey R, et al. Endothelial ultrastructural alterations of intramuscular capillaries in infantile mitochondrial cytopathies: “Mitochondrial angiopathy.” Neuropathology 2012; 32: 617–27
97. Sakuta R, Nonaka I. Vascular involvement in mitochondrial myopathy. Ann Neurol 1989; 25: 594–601
© 2013 by American Association of Neuropathologists, Inc.