There are four major categories of idiopathic inflammatory myopathy: dermatomyositis (DM), polymyositis (PM), immune-mediated necrotizing myopathy (IMNM), and inclusion body myositis (IBM), which are clinically, histologically, and pathogenically distinct (Tables 33-1,33-2,33-3).^1–12 These myositides may occur in isolation or in association with cancer, various connective tissue diseases (overlap syndromes), and autoantibodies. Other less common idiopathic myositides (i.e., granulomatous and myositis associated with infections) will also be discussed in this chapter. DM and IBM are rather homogeneous clinically and histologically. On the other hand, what has been called “PM” in the literature is likely a heterogeneous group of disorders. It is important to emphasize that not all myopathies with inflammation are classified as “inflammatory myopathies.” In this regard, various muscular dystrophies (e.g., congenital, facioscapulohumeral, and dysferlinopathies) may be associated with profound inflammation and are not uncommonly misdiagnosed as PM.

There are a few reports of idiopathic inflammatory myopathy occurring in parents, children, and siblings of affected patients, suggesting a genetic predisposition to developing these disorders, possibly secondary to inherited human leukocyte antigens (HLA) haplotypes.^13–15 There are hereditary forms of inclusion body myopathy, but with rare exceptions, the muscle biopsies in these cases lack inflammation, and the clinical phenotype (i.e., age of onset and pattern of weakness) is different from sporadic IBM.

The annual incidence of the inflammatory myopathies as a whole has ranged between 0.1 and 1 per 100,000 person years,^3,5,16–21 with more recent studies suggesting the incidence may be greater than four cases per 100,000²² with prevalence in the range of 14 to 32 per 100,000.^5,22–25 However, defining the actual incidence and prevalence of the individual myositides has been limited by the different diagnostic criteria employed in various epidemiological studies. Most published papers regarding epidemiology and treatment of DM and PM have used Bohan and Peter criteria (Table 33-4).^26–28 PM will be overdiagnosed with Bohan and Peter criteria. These criteria were fine in 1975, but these criteria do not require a muscle biopsy and the only feature that distinguishes PM from DM is the presence of a rash in DM. Further, the biopsy abnormalities as listed are nonspecific (except for perifascicular atrophy—a finding specific for DM, but not seen in PM) and do not help in distinguishing PM from DM or for that matter any myopathy with necrosis, including muscular dystrophies. Importantly, the Bohan and Peter histological criteria do not take into account the advances in histopathology and the distinct diagnoses of IBM and IMNM. This can have implications in regard to treatment strategies and prognosis as we will discuss.²⁹

Criteria for diagnosis of the various inflammatory myopathies need to take into account the advances in understanding of the pathogeneses of these disorders. We emphasize that DM is not simply PM with a rash (or the converse: PM is not DM without a rash). Furthermore, IBM is not PM with inclusions (or the converse: IBM is not PM with inclusions). For this reason, revised criteria for the various idiopathic inflammatory myopathies have been devised to take into account the recent advancements in the field (Tables 33-2,33-3).^30–32 For definitive histopathological diagnosis of PM in the biopsy, some require endomysial infiltrates composed of CD8+ T cells and macrophages invading non-necrotic muscle fibers that express major histocompatibility-1 (MHC-1) antigen.^4,30–32 However, the sensitivity of this finding is low, and this biopsy feature is not diagnostic for PM, as it also is seen in IBM and rarely in dystrophies. Likewise, perivascular, perimysial, and endomysial inflammatory infiltrates are nonspecific findings and can be found in DM, PM, IBM, dystrophies, toxic, and metabolic myopathies. With the caveats noted above, we will begin our discussion of the individual inflammatory myopathies.

CLINICAL FEATURES

DM can present at any age, including infancy. Similar to most other autoimmune disorders, there is an increased incidence of DM in women compared to men.^{11,16–19,33} Although the pathogenesis of childhood and adult DM is presumably similar, there are important differences in some of the clinical features and associated disorders. Weakness can develop rather acutely (over days or several weeks), or insidiously (over months).^17,34,35 Proximal leg and arm muscles as well as neck flexors are usually the earliest and most severely affected muscle groups. Thus, the earliest patient complaints are often difficulty lifting their arms over their heads, climbing steps, and arising from chairs. Distal muscles are also involved. Children are more likely to present with an insidious onset of muscle weakness and myalgias that are often preceded by fatigue, low-grade fevers, and a rash. Dysphagia occurs in approximately 30% of patients with DM probably due to involvement of oropharyngeal and esophageal muscles.³ Speech, chewing, and swallowing difficulties can arise secondary to involvement of the masseter muscle. We have even seen speech difficulties as a result of involvement of the pharyngeal, laryngeal, and the tongue muscles. Sensation is normal, and muscle stretch reflexes are preserved unless a severe degree of weakness has developed.

DM is usually diagnosed earlier than other forms of myositis because of the characteristic rash, which typically accompanies or precedes the onset of muscle weakness.^1,3,34,36 However, the rash can develop years after the onset of weakness, which could lead to an erroneous diagnosis of PM. Some patients have the characteristic rash but never develop weakness (the so-called amyopathic DM or DM sine myositis).^36,37 Rare patients do not have an appreciable rash at the time they present with weakness. We have seen some patients with histopathological features characteristic of DM who have developed the rash months or years after onset of weakness or not at all (adermatopathic DM or DM sine dermatitis). These patients would be erroneously classified as PM using Bohan and Peter criteria.

The classical skin manifestations include a purplish discoloration of the eyelids (heliotrope rash) often associated with periorbital edema and a papular, erythematous rash over the knuckles (Gottron’s papules) (Fig. 33-1). In addition, an erythematous, macular, sun-sensitive rash may appear on the face, neck and anterior chest (V-sign), shoulders and upper back (shawl sign), hips (holster sign), and extensor surfaces of elbows, knuckles, knees, and malleoli (Gottron’s sign). The nail beds often have dilated capillary loops occasionally with thrombi or hemorrhage. The skin lesions can be subtle at times and difficult to appreciate in individuals who are darker skinned—another common reason for misdiagnosing patients with PM rather than DM.

Subcutaneous calcifications occur in 30–70% of children, but in our experience these are less common in adults (Fig. 33-2).^38,39 These lesions tend to develop over pressure points (buttocks, knees, and elbows) and can be complicated by ulceration of the overlying skin. Once the calcinosis appears, treatment is very difficult. Colchicine, probenecid, warfarin, and phosphate buffers have been tried with limited success. Surgery may be performed, but the lesions may recur or worsen.

ASSOCIATED MANIFESTATIONS

Cardiac

Conduction defects, arrhythmias, ventricular, and septal wall motion abnormalities, and reduced ejection fractions may be seen on electrocardiograms, echocardiography, and radionucleotide scintigraphy.^{34,35,40–44} Nevertheless, most patients do not develop any cardiac symptoms. However, pericarditis, myocarditis, and congestive heart failure can occasionally develop secondary to involvement of cardiac muscle and may be lethal.^34,42

Pulmonary

Interstitial lung disease (ILD) complicates approximately 10–20% of patients with DM.^34,45–48 Rarely, patients develop bronchiolitis obliterans with organizing pneumonia. ILD manifests clinically as dyspnea and nonproductive cough. It can begin abruptly or insidiously and even precede the development of the characteristic rash and muscle weakness. Chest radiographs reveal a diffuse reticulonodular pattern with a predilection for involvement at the lung bases. A diffuse alveolar pattern or ground-glass appearance is seen in the more fulminant cases.⁴⁵ A restrictive defect with reduced forced vital capacity and decreased diffusion capacity are evident on pulmonary function tests. Antibodies directed against t-histidyl transfer RNA synthetase, the so-called Jo-1 antibodies, are present in at least 50% of ILD cases associated with inflammatory myopathies.^49–51 A less common pulmonary complication is ventilatory muscle weakness, but it does occur. Furthermore, aspiration pneumonia can be a complication of oropharyngeal and esophageal weakness.

Gastrointestinal

Involvement of the skeletal and smooth muscles of the gastrointestinal tract can lead to dysphagia, aspiration, and delayed gastric emptying. Vasculopathy affecting the gastrointestinal tract is a serious complication that appears to be much more common in juvenile DM compared to adult DM. The vasculopathy can result in mucosal ulceration, perforation, and life-threatening hemorrhage.

Joints

Arthralgias of large and small joints with or without arthritis are common. Joint and muscle pain often eases when the limbs are flexed, and this can lead to the formation of flexion contractures across the major joints. This emphasizes the importance of early physical therapy and range of motion exercises to prevent contractures from developing. Flexion contractures at the ankles leading to toe walking are a common early finding in childhood DM.

Vasculopathy

A vasculopathy affects the skin, muscle, and gastrointestinal system. Rarely, massive muscle infarction can lead to myoglobinuria and acute renal tubular necrosis.

Malignancy

There is an increased incidence of cancer ranging from 6–45% in DM.^{27,28,34,35,52,53} The association with cancer has not been demonstrated in juvenile DM and the increased risk is predominantly seen in adults over the age of 40 years. Although women are more likely to develop DM than men, the risk of malignancy is equal in both sexes. Most malignancies are identified within 2 years of the presentation of the myositis. The clinical severity of rash or muscle weakness does not appear to correlate with the presence or absence of a neoplasm. Treatment of the underlying malignancy sometimes results in improvement of muscle strength.

We perform a comprehensive history and annual physical examination with breast and pelvic examinations for women and testicular and prostate examinations for men to search for an underlying malignancy. In addition, we obtain a complete blood count (CBC), routine blood chemistries, urinalysis, and stool specimens for occult blood. Computerized tomographic (CT) scans of the chest, abdomen, and pelvis and mammography are also ordered. Colonoscopy should be done on all patients over the age of 50 years or in those who have attributable gastrointestinal symptoms (e.g., abdominal pain, constipation, or blood in the stool).

LABORATORY FEATURES

Necrosis of muscle fibers usually leads to increased serum creatine kinase (CK), aldolase, myoglobin, lactate dehydrogenase, aspartate aminotransferase (AST), and alanine aminotransferase (ALT) levels. Serum CK is the most sensitive and specific marker for muscle damage and is elevated in at least 90% of patients with DM.^27,34,35 However, serum CK levels do not correlate with the severity of weakness. The CK level can be normal even in individuals who are markedly weak, particularly in childhood DM, in patients with slow, insidious disease, and in those with little residual muscle mass. In approximately 10% of cases with a normal CK, the aldolase level is elevated.^54–56 Erythrocyte sedimentation rate (ESR) is usually normal or only mildly elevated and is not a reliable indicator of disease severity.

Antinuclear antibodies (ANAs) are detected in 24–60% of patients with DM.^34,50,51 These antibodies are much more common in patients with overlap syndromes (to be discussed later). Some patients have the so-called “myositis-associated antibodies” or “myositis-specific antibodies” (MSAs) (Table 33-5).^{5,34,50,51,57–74} Autoantibodies to various aminoacyl tRNA synthetases (ARSs) constitute the majority of the MSA of which Jo-1 antibodies (directed against histidyl t-RNA synthetase) are the most common. Jo-1 are demonstrated in as many as 20% of patients with inflammatory myopathy and are strongly associated with ILD.^5,49–51 Some patients have ILD but no myositis, so one can argue that these antibodies are not necessarily specific for myositis. The other antisynthetases are much less common and are each found in <23% of patients with inflammatory myopathy. It has been suggested that the presence of Jo-1 antibodies is associated with only a moderate response to treatment and a poor long-term prognosis.^59–61 However, there has not been a prospective study of treatment outcomes comparing patients with myositis associated ILD with Jo-1 antibodies with patients without these antibodies. Nevertheless there is a constellation of features that constitute the so-called “antisynthetase syndrome” including ILD, myositis, Raynaud’s syndrome, arthritis, and mechanic hands which can be useful in diagnosis.

Mi-2 antibodies are found in 15–20% of patients with DM. Mi-2 is a 240-kD nuclear protein of unknown function. The Mi-2 antibodies are typically associated with an acute onset, a florid rash, a good response to therapy, and a favorable prognosis.^{5,50,51,59–61} However, again it is not known if patients with DM patients with Mi-2 antibodies respond differently than DM without the antibody.

Antibodies directed against melanoma differentiation-association protein 5 (anti-MDA5), also known as anti-CADM-140 antibodies, are found in 10–20%of DM patients and up to 65% of patients with clinically amyopathic DM and are associated with rapidly progressive ILD, particularly in Asians.^62–67 Anti-MDA-5 antibody levels closely correlate with the severity of skin ulcerations, ILD, and disease prognosis.

Autoantibodies targeting transcriptional intermediary factor 1-γ (TIF1-γ), also known p155 antibodies, are found in adult cancer-associated DM with an 89% specificity and 70% sensitivity.⁷³ Antibodies directed against nuclear matrix protein NXP-2 (also known as MORC3) have been found in as many as 17% of patients with DM and are also associated with cancer.⁷³

Magnetic resonance imaging (MRI) can provide information on the pattern of muscle involvement by looking at the cross-sectional area of axial and limb muscles.^74–79 MRI may demonstrate signal abnormalities in affected muscles secondary to inflammation and edema or replacement by fibrotic tissue. However, the changes on MRI are not usually specific for myositis. Some have advocated MRI as a method to guide which muscle to biopsy.⁷⁸ However, we have found that MRI usually adds little to a good clinical examination and EMG in defining the pattern of muscle involvement and determining the muscle for biopsy.

ELECTROPHYSIOLOGICAL FEATURES

The characteristic electromyography (EMG) abnormalities observed in patients with myositis include (1) increased insertional and spontaneous activity with fibrillation potentials, positive sharp waves, and occasionally pseudomyotonic discharges (e.g., decrescendo waves of positive waves that do not wax or wane in frequency and amplitude) or complex repetitive discharges; (2) small-duration, low-amplitude, polyphasic motor unit action potentials (MUAPs); and (3) MUAPs that recruit early but at normal frequencies.⁸⁰ Recruitment may be decreased (fast firing MUAPs) in advanced disease, if there is marked loss of muscle fibers. Decreased insertional activity may be seen in chronic disease secondary to fibrosis. In addition, large-duration, polyphasic MUAPs may also be evident later in longstanding disease due to muscle fiber splitting and regeneration rather than a superimposed neurogenic process.

The degree of abnormal spontaneous EMG activity reflects the ongoing disease activity. EMG can be used to assist determining which muscle to biopsy in patients with only mild weakness. In addition, EMG may also be useful in the assessment of previously responsive patients with myositis who become weaker by differentiating an increase in disease activity from weakness secondary to type 2 muscle fiber atrophy from disuse or chronic steroid administration. Abnormal insertional and spontaneous activity is expected in active myositis, while isolated type 2 muscle fiber atrophy is not associated with such abnormal activity on EMG. Along these lines, it is our opinion that a multifocal or diffuse pattern of abnormal insertional and spontaneous activity without obvious changes in MUAP morphology or recruitment is much more likely to represent an acute myopathy, like DM, than a neurogenic disorder.

HISTOPATHOLOGY

The pathological process is multifocal, and the frequency and severity of histological abnormalities can vary within the muscle biopsy specimens. The pathognomonic histological feature is perifascicular atrophy (Fig. 33-3A), although this is a late finding and in our experience is found in <50% of adult-onset cases (it is somewhat more frequent in juvenile-onset DM). The perifascicular area contains small regenerating and degenerating fibers. Oxidative enzyme stains highlight the microvacuolation within these fibers. Combined COX/SDH stain may demonstrate COX-negative/SDH-positive staining perifascicular muscle fibers (Fig. 33-3B). Scattered necrotic fibers and much less frequently, wedge-shaped microinfarcts may be evident. Even though DM is an inflammatory myopathy, inflammatory cell infiltrates are not evident with routine histochemistry in some patients. The inflammatory infiltrate is composed primarily of macrophages, B cells, and CD4+ cells in the perivascular and perimysial regions around blood vessels (perivascular).^32,81 These CD4+ cells are mainly plasmacytoid dendritic cells (PDCs) and not T-helper cells as they are often CD3 negative.⁸² Importantly, in contrast to PM and IBM (discussed later), invasion of non-necrotic fibers is not prominent. Immunohistochemistry (IHC) staining demonstrate that muscle fibers express MHC-1 antigen, STAT1, and various interferon-α/β inducible proteins, including myxovirus resistance 1 (MxA) and ISG15 on the sarcolemma, particularly in the perifascicular regions (Fig. 33-4), and can be seen even before the development of perifascicular atrophy.^82,83 However, there is no overexpression of interferon-gamma inducible proteins on muscle fibers.

DM is associated with a reduction in the capillary density (number of capillaries per area of muscle) and compensatory dilation of the remaining small vessels.⁸⁴ One of the earliest demonstrable histological abnormalities in DM is deposition of the C5b-9 complement membrane attack complex (MAC) around small blood vessels (Fig. 33-5).^84–86 Deposition of MAC precedes inflammatory cell infiltration and other structural abnormalities (e.g., perifascicular atrophy) in the muscle on light microscopy and is relatively specific for DM.⁸⁴ Other complement components (C3 and C9) and immunoglobulins (IgM and less often IgG) are also deposited on or around the walls of intramuscular blood vessels.⁸⁷ These observations have led to the hypothesis that DM is caused by deposition of immunoglobulins on capillaries, subsequent activation of complement, and MAC-induced necrosis of the vessels, which then lead to ischemic damage of muscle. However, as discussed in section “Pathogenesis”, this hypothesis is purely speculative. In addition to expression of MxA on muscle fibers, MxA also is expressed on capillaries in DM (Fig. 33-4).

Electron microscopy (EM) reveals small intramuscular blood vessels (arterioles and capillaries) with endothelial hyperplasia, microvacuoles, and tubuloreticular cytoplasmic inclusions.^12,88

PATHOGENESIS

Immunological studies and other histological features on muscle biopsies have suggested that DM may be at least in part a humorally/complement mediated microangiopathy, although this is far from proven.⁸⁹ Although the presence of MAC is well established, its frequency is not, and this is important with regard to the possibility of varied mechanisms of disease and distinct subtypes of DM. Its presence on blood vessels could be due to either immune complex deposition, or complement activation by either the classical antibody-mediated or alternative pathways. Even classical pathway activation, which is antibody dependent, can still be relatively antigen nonspecific; some IgM antibodies are highly polyclonal, binding with low avidity to many self-antigens. The specificity of MAC presence is also in question, as it is present in abnormal vascular tissue (e.g., atherosclerotic coronary arteries). It may be that the microvasculature is damaged by some other mechanism (e.g., interferon- or other cytokine-related toxicity), and the deposition of immunoglobulins and complement on the damaged vascular tissue might be a secondary phenomenon. That some individuals have developed DM with hereditary complement deficiencies argues against primary destruction of capillaries by complement and MAC.^90,91

The microangiopathy has been postulated to cause ischemic damage and occasionally infarction of muscle fibers. It has been suggested that the perifascicular atrophy is the result of hypoperfusion to the watershed region of muscle fascicles. However, it has never been demonstrated that the perifascicular region is indeed the watershed area in muscle fibers and that perifascicular fibers are more prone to ischemic damage.⁹² Perifascicular atrophy and endomysial capillary MAC deposition were found, in one study, to be inversely correlated,⁸⁵ and another study found no correlation between perifascicular atrophy and capillary depletion.⁸⁴ Furthermore, perifascicular atrophy has not been reported in vasculitis, a condition with known muscle ischemia and infarction, nor has perifascicular atrophy been found in experimental models of skeletal muscle ischemia.⁹² In another model of ischemic myopathy, resulting from microarterial embolization with particles 20–80 m in diameter, the pathological changes were located centrally within fascicles and the perifascicular regions were instead preferentially spared.⁹³, Finally, perifascicular atrophy is not evident in ischemic muscle in animal models when muscle is rendered ischemic from vasculitis or other small vessel injury.

Gene microarray studies of biopsied muscle tissue demonstrate an increased expression of genes induced by type 1 interferons.^82,94,95 Although this is not specific, it is compatible with the hypothesis of a viral infection triggering the autoimmune attack as interferons have a well-defined role in antiviral innate immunity. However, there are other possibilities. Type 1 interferons (i.e., interferon-α and interferon-β) are synthesized by PDCs in response to a serum factor(s) containing immune complexes of antibody, double-stranded DNA, or RNA viruses. Abundant PDCs are evident in the muscle biopsies of patients with DM.⁸² PDCs are CD4+ and comprise a large component of the inflammatory cell infiltrate in DM. These CD4+ cells were originally thought to be CD4+ T-helper cells, but it was subsequently demonstrated that most are CD3-. Therefore, these CD4+ cells are predominantly PDCs and not lymphocytes. Increased expression of interferon-α/β inducible protein MxA is evident on blood vessels and muscle fibers (with a predilection for the perifascicular fibers). Interestingly, one postulated function of MxA is to form tubuloreticular inclusions around RNA viruses. These inclusions have the same morphology as the tubuloreticular inclusions seen on EM in blood vessels in DM. Using immunoelectron microscopy, MxA was demonstrated within inclusions in vessels in DM muscle biopsies.⁸² Interestingly, increased expression of type 1 interferon regulated genes are also evident in the peripheral blood and on skin biopsies of patients with active DM, similar to what has been described in systemic lupus erythematosus (SLE). Further, expression levels in the blood appear to correlate with disease activity. We suspect that dysregulated interferon-α/β production plays a major role in the pathogenesis of DM and could be directly toxic to the small blood vessels and muscle fibers themselves.

PROGNOSIS

In the absence of malignancy, prognosis is generally favorable in patients with DM. Poor prognostic features are increased age, associated ILD, cardiac disease, and late or previous inadequate treatment with 5-year survival rates ranging from 70% to 93%.^35,59,96,97

PM, as reported in the literature, is likely to be a heterogeneous group of disorders rather than a distinct entity. A major source of debate among clinicians who primarily take care of patients with PM (e.g., neurologists and rheumatologists) is the criteria for diagnosing PM. The most commonly employed criteria were developed by Bohan and Peter in 1975 (Table 33-4),^26,27 but these do not take into account advancements in our understanding of the immunopathogenesis of the various inflammatory myopathies or even the existence of IBM and IMNM. Revised criteria for the various idiopathic inflammatory myopathies have been proposed (Tables 33-2,33-3). For definitive histopathological diagnosis of PM, these criteria require CD8+ T cells invading non-necrotic muscle fibers that express MHC-1 antigen.^9,30,31 Even so, this biopsy feature is not diagnostic for PM, as it also is seen in IBM and rarely in dystrophies. Further, mononuclear cell invasion of non-necrotic muscle fibers is uncommon in suspected cases of PM, and some argue that it is not necessary for the diagnosis of PM.^58,98 More frequently on biopsy we appreciate perivascular/perimysial inflammatory cell infiltrates or endomysial inflammatory cells, but no actual invasion on non-necrotic muscle fibers.^11,31 Whether or not these cases represent “PM,” with the absence of CD8+ T cells invading non-necrotic muscle fibers or a distinct type of inflammatory myopathy, is unclear. Such perivascular, perimysial inflammation is common, particularly in patients with overlap myositis, but can be seen in DM and IBM and, occasionally, in dystrophies.

For the various reasons listed above, it is impossible to extract from the literature the true incidence and prognosis of PM or its subtypes and the associated laboratory abnormalities, medical conditions (e.g., connective tissue disorder [CTD], ILD, myocarditis, and cancer). We need prospective studies using contemporary clinical, laboratory, and histopathological criteria for PM to address these issues. Nevertheless, we will summarize the available literature regarding “PM.”

CLINICAL FEATURES

PM generally presents in patients over the age of 20 years. Unlike DM, idiopathic PM in absence of an underlying connective tissue disease is rare in childhood in our experience. As in DM and other autoimmune disorders, PM is more prevalent in women.^{16–28,34,35} The diagnosis of PM is often delayed compared to DM. As with DM, patients present with symmetric proximal arm and leg weakness that typically develops over several weeks or months. Distal muscles may also become involved but are not as weak as the more proximal muscles. Muscle pain and tenderness are frequently noted but these are not the primary symptoms—weakness is the primary complaint. Approximately one-third of patients complain of swallowing difficulties. Mild facial weakness occasionally may be demonstrated on examination. Sensation is normal and muscle stretch reflexes are usually preserved.

ASSOCIATED MANIFESTATIONS

The cardiac and pulmonary complications of PM are reportedly similar to that described in the DM section. Myositis with secondary congestive heart failure or conduction abnormalities occur in up to one-third of patients, but again histopathological confirmation of definite PM using more up-to-date criteria is lacking in most of these studies.^{34,35,40–44} Anti-signal recognition particle (SRP) antibodies have been associated with myocarditis and were felt to be specific for PM, although the histopathology is more often that of a necrotizing myopathy.^99,100 In studies of SRP-myositis in which detailed immunohistochemistries were performed, the biopsies were not suggestive of PM (i.e., inflammatory infiltrate was scant), but rather revealed features suggestive of a microvasculopathy.^82,101 ILD has been reported to occur in at least 10% of patient with PM, with at least half having Jo-1 antibodies.^{34,45–48,50,51,61} Muscle biopsies from patients with Jo-1 antibodies demonstrated perimysial abnormalities (fragmentation and increase staining with alkaline phosphatase of the perimysial connective tissue, atrophy and degeneration of perifascicular muscle fibers, and perivascular/perimysial inflammation without endomysial inflammatory cells invading non-necrotic muscle fibers.^10,101 This suggests to us that anti-SRP and anti–Jo-1 myositis cases are probably distinct from idiopathic PM.

Polyarthritis has been reported in as many as 45% of patients with PM at the time of diagnosis.³⁵ The risk of malignancy with PM seems to be lower than that seen in DM, but is slightly higher than expected in the general population.^28,35,52,53

LABORATORY FEATURES

Serum CK level is elevated fivefold or more in most PM cases.^27,28,34,35 Unlike DM and IBM (to be discussed later) in which the CK can be normal, the serum CK should be elevated in active PM. Serum CK can be useful in monitoring response to therapy, but only in conjunction with the physical examination, as the CK level does not necessarily correlate with the degree of weakness. ESR is normal in most patients and does not correlate with disease activity or severity.

Positive ANAs are reportedly present in 16–40% of patients with PM.^27,34,35,50 However, again the exact relationship of ANAs and CTD in patients with histologically defined PM is unclear. The relationships of various MSAs to PM were previously addressed (Table 33-5).

MRI may demonstrate T2 signal abnormalities in affected muscles secondary to inflammation and edema or T1 abnormalities as a result of replacement by fibrotic tissue (Fig. 33-6).^74–78

ELECTROPHYSIOLOGICAL FEATURES

EMG is usually abnormal in PM with increased insertional and spontaneous activity, small polyphasic MUAPs, and early recruitment.⁸⁰ These abnormal features do not distinguish PM from other inflammatory myopathies or myopathies with muscle membrane instability.

HISTOPATHOLOGY

The histological features of PM are distinct from DM. The predominant histological features in PM are variability in fiber size, scattered necrotic and regenerating fibers, and inflammatory cell infiltrates. However, as mentioned previously, the specific characteristics of this inflammatory cell infiltrate have been the subject of recent debate. Small studies of PM reported that muscle biopsies demonstrate CD8+ T cells and macrophages invading non-necrotic muscle fibers expressing MHC-1 antigen (Fig. 33-7).^31,32,81,102 Subsequently, some have argued that this histopathological feature is required for the diagnosis of definite PM.^4,30,31 However, other authorities argue that invasion of non-necrotic muscle fibers is not necessary and perivascular, perimysial, or endomysial inflammation without actual invasion of non-necrotic muscle fibers can suffice for the diagnosis of PM in the proper clinical context.^58,98 In our opinion, demonstrating invasion of non-necrotic endomysial muscle fibers by T cells is very helpful in making a diagnosis of PM on histopathological grounds, as the sole findings of perivascular, perimysial, and even endomysial inflammatory cell infiltrates can be seen in DM, some dystrophies, and rhabdomyolysis from metabolic and toxic myopathies. Importantly, invasion of non-necrotic muscle fibers is not diagnostic for PM and is actually more commonly seen in IBM as will be discussed later. That said, invasion of non-necrotic muscle fibers is not mandatory to make a clinical diagnosis of PM in the appropriate situation.

The endomysial inflammatory cells consist primarily of activated CD8+ (cytotoxic), alpha, and beta T cells and macrophages.^32,81,102 Rare cases of PM with CD4- and CD8-gamma/delta T-cell infiltrates have been reported.^103–105 The T-cell receptors of endomysial T cells have an oligoclonal pattern of gene rearrangements and a restricted motif in the CD3R region, suggesting that the immune response is antigen specific.^106,107 Further, there are many myeloid dendritic cells in the endomysium that appear to surround non-necrotic muscle fibers and may serve to present antigens to cytotoxic T cells. Although B cells are rare, plasma cells are common in the endomysium and likely account for the increased expression of immunoglobulin genes on microarray experiments.¹⁰⁸ There is also evidence of oligoclonal pattern of gene rearrangements in plasma cells in PM muscle biopsies. Unlike DM, MAC, complement, or immunoglobulins are not deposited on the microvasculature in PM.

PATHOGENESIS

PM is believed to be the result of an HLA-restricted, antigen-specific, cell-mediated immune response directed against muscle fibers. The trigger of this autoimmune attack is not known, but viral infections have been speculated. However, there is no conclusive evidence supporting this hypothesis.¹⁰⁹ MHC-1 molecules on the surface of cells usually express endogenous self-peptides rather than viral particles. Neither viral proteins nor DNA have been identified in muscle fibers. Thus, the autoimmune response may be directed against endogenous self-antigens rather than processed viral antigens. Nonetheless, a viral infection could indirectly trigger an immune response secondary to antigenic mimicry with muscle proteins, altering the expression of proteins on the surface of muscle fibers such that these become antigenic, or by the loss of physiological self-tolerance. Myositis may complicate human immunodeficiency virus (HIV) and human T-lymphocyte virus-1 (HTLV-1) infections.¹¹⁰ In these cases, the myositis appears to be the result of such indirect triggering of the immune response against muscle fibers.

The cytotoxic T cells appear to destroy muscle fibers via the perforin pathway. These autoinvasive T cells contain perforin granules oriented next to the sarcolemma of muscle fibers.¹¹¹ Upon release of these granules by exocytosis, pore formations are induced on the sarcolemma, leading to osmolysis of muscle fibers.

DIFFERENTIAL DIAGNOSIS

A diagnosis of PM relies on a thorough search to exclude other causes of weakness (Table 33-6).^29,112 A detailed clinical examination of an appreciation of the pattern of weakness can help differentiate IBM and muscular dystrophies with inflammation from PM. Serum CK should be elevated in PM, while it is normal in patients with “fibromyalgia” and polymyalgia rheumatica and can be normal in IBM. Skeletal muscle MRI is often interpreted as showing “myositis.” However, these increased signal abnormalities are not specific and can be seen in dystrophies, rhabdomyolysis from toxic medications (e.g., statins) metabolic myopathy, and muscle infarcts from various causes (e.g., vasculitis and diabetic vasculopathy). The specific pattern of muscle involvement and extensive fatty replacement in the absence of edematous changes on MRI scans would be helpful, suggesting a dystrophy as opposed to PM. EMG can be useful, as the presence of diffuse myotonic discharges should lead to the consideration of proximal myotonic myopathy or late-onset acid maltase deficiency—conditions that we have seen misdiagnosed as PM.

Importantly, the diagnosis of PM requires a muscle biopsy. It is important to look for histopathological features that would suggest IBM (e.g., rimmed vacuoles, inclusions, ragged red fibers, etc.). However, the absence of these findings does not exclude the diagnosis of IBM. Muscle biopsy is essential to look for features that might suggest a dystrophy, metabolic myopathy such as acid maltase deficiency, or necrotizing myopathy.

PROGNOSIS

Most patients with PM improve with immunosuppressive therapies but usually require life-long treatment.^34,59,96 Some retrospective studies suggest that PM does not respond to immunosuppressive agents as well as DM. However, interpretation of the results of these retrospective series is difficult, as the diagnosis of PM was usually made based on the Bohan and Peter criteria rather than on more up-to-date criteria based on strict clinical and histological criteria.

The term “overlap syndrome” is applied when DM or PM is associated with other well-defined CTDs such as scleroderma, mixed connective tissue disease (MCTD), Sjögren’s syndrome, SLE, or rheumatoid arthritis.^1,3,4,11 In our experience^82,83,94 and others,^11,31 the muscle biopsies in patients with overlap syndrome resemble DM, a necrotizing myopathy, or are associated with nonspecific (e.g., perivascular and perimysial) inflammatory cell infiltrates as opposed to PM (at least if defined by CD8+ cells invading non-necrotic muscle fibers). These features correspond to what some have termed myopathy with perimysial pathology.¹⁰ The prognoses in these patients are related in part to the underlying CTD. Retrospective series of patients that suggest that myositis associated with overlap syndromes is more responsive to immunosuppressive treatment than isolated DM and PM, but again prospective studies are lacking.^{10,34,49,59,96}

SCLERODERMA

Weakness is common in scleroderma. Most patients have normal serum CKs and EMG, while muscle biopsies demonstrate only mild variability in fiber size with atrophy of type 2 muscle fibers and perimysial fibrosis. However, 5–17% of patients with scleroderma have myositis which can occur in either of its two major forms—progressive systemic sclerosis or CREST (Calcinosis, Raynaud’s phenomena, Esophageal dysmotility, Sclerodactyly, Telangiectasia) syndrome (Fig. 33-8).^35,113–116 Patients with scleroderma myositis may have mildly increased serum CK levels and irritable and myopathic EMGs. Detailed descriptions of the immunohistopathology on muscle biopsies are lacking, and therefore it is difficult to ascertain if these have features of DM or PM.

Most patients with CREST syndrome have anticentromere antibodies, while anti-Scl-70 antibodies are common in patients with progressive systemic sclerosis. Some patients with scleroderma myositis have anti-PMScl (also called anti-PM-1) antibodies.^51,117

SJÖGREN’s SYNDROME

Sjögren’s syndrome is characterized by dryness of the eyes and mouth (sicca syndrome) and other mucosal membranes. Muscle pain and weakness are common in Sjögren’s syndrome, but true myositis is rare. Muscle weakness is usually due to disuse atrophy secondary to arthritis and pain. Nonetheless, myositis can occur with Sjögren’s syndrome.^35,118–120 About 90% of patients have ANAs directed against ribonucleoproteins, specifically SS-A (Ro) and less commonly SS-B (La) antibodies.

SYSTEMIC LUPUS ERYTHEMATOSUS

SLE is an autoimmune disorder affecting multiple organ systems. As with other CTDs, weakness is not unusual in SLE but is most often the result of disuse atrophy. Nevertheless, myositis can occur with SLE.^{35,82,121,122}

Most patients with SLE have positive ANA titers that are directed against native DNA (highly specific for SLE) and ribonuclear proteins (RNPs). The anti-RNP antibodies are present in less than half of patients with SLE and include anti-SS-A and anti-SS-B (also present in Sjögren’s syndrome), anti-U1 RNP (also present in MCTD), and anti-Sm (specific for SLE).

Of note, gene expression studies in peripheral blood of patients with SLE demonstrated an upregulation of type 1 interferon-inducible genes, similar to what is seen in gene expression studies of muscle biopsies in DM.⁸² In this regard, MxA is highly expressed in both SLE blood and DM muscle. Both disorders are also associated with tubular reticular inclusions in endothelial cells on EM. Thus, DM and SLE likely share a similar pathogenic basis with abnormalities involving the innate immune system.

RHEUMATOID ARTHRITIS

The most common etiology of weakness in RA is type 2 muscle fiber atrophy from chronic steroids or disuse secondary to arthritis, but myositis can infrequently occur.³⁵

MIXED CONNECTIVE TISSUE DISEASE

Patients with MCTD have clinical features of scleroderma, SLE, rheumatoid arthritis, and myositis.¹²³ In terms of the myositis, DM is reported more commonly than PM.^{35,82,113,123} This is certainly our experience. Necrotizing myopathy can also complicate MCTD. High titers of anti-U1 RNP antibodies are common in MCTD but are nonspecific, as these can also be detected in SLE.

CLINICAL FEATURES

IBM is characterized clinically by the insidious onset of slowly progressive proximal and distal weakness, which generally develops after the age of 40 years (and usually after 50 years) (Tables 33-1,33-3).^{1–3,9,124–132} IBM appears to be the most common myopathy (apart from sarcopenia of aging) in patients over the age of 50 years. The slow progressive nature of the myopathy probably accounts in part for the delay in diagnosis that averages 6–7 years after the onset of symptoms. Men are much more commonly affected than women, in contrast to the female predominance seen in DM and PM.

The clinical hallmark of IBM is early weakness and atrophy of the quadriceps, flexor forearm muscles (i.e., wrist and finger flexors) (Fig. 33-9), and ankle dorsiflexors.^{1–3,9,126–128} This pattern of weakness is present in as many as two-thirds of patients with IBM, but not all.¹³² With manual muscle testing, the MRC grades of the finger and wrist flexors (in particular the deep finger flexors such as the flexor pollicis longus) are usually lower than those of the shoulder abductors, and the muscle scores of the knee extensors and ankle dorsiflexors may be the same or lower than those of the hip flexors in patients with IBM.^2,129–132 In contrast, the proximal muscles (shoulder abductors and hip flexors) are usually weaker than distal muscle groups by manual muscle testing grades in DM and PM. In addition, muscle involvement in IBM is often asymmetric, in contrast to the symmetrical involvement in DM and PM. The asymmetric involvement of muscle, not uncommonly, leads to the misdiagnosis of amyotrophic lateral sclerosis (ALS). However, the muscle groups affected early are different in IBM compared to ALS. Again in IBM, there is an atrophy of the flexor forearm compartment, but the hand intrinsics (thenar and hypothenar eminence) are spared, in contrast to ALS in which atrophy in the upper limbs usually is first seen in the hand intrinsics. The presence of slowly progressive, asymmetric, quadriceps and wrist/finger flexor weakness, and atrophy in a patient over 50 years of age strongly suggests the diagnosis of IBM even in the absence of histological confirmation.^2,129–132 Although slowly progressive, IBM is very debilitating. Longitudinal studies have reported that 37% of patients used a wheelchair after 14 years,¹²⁸ while others have reported 47% of IBM patients being completely confined to a wheelchair after only 12 years.¹²⁷

Swallowing difficulties develop in up to 60% of patients due to esophageal and pharyngeal muscle involvement. This can lead to weight loss or aspiration. In severe cases, cricopharyngeal myotomy may be beneficial.^126,133,134 We have followed a number of patients in whom dysphagia was the presenting feature of the disease. Only after following patients for several years did they develop weakness in the extremities that are more characteristic of IBM. Mild facial weakness is evident in one-third of cases.^2,126 Rare patients may have severe facial diplegia.¹³⁵ In keeping with other inflammatory myopathies, neck flexor weakness is the rule, but some patients manifest with atypical features such as with head drop or bent spine syndrome/camptocormia owing to severe paraspinal muscle involvement.^136,137 Most patients have no sensory symptoms, but as many as 30% have evidence of a generalized sensory peripheral neuropathy on clinical examination and electrophysiological testing.² Muscle stretch reflexes are normal or slightly decreased. In particular, the patellar reflexes are lost early.

ASSOCIATED MANIFESTATIONS

Unlike DM and PM, IBM is not associated with myocarditis, lung disease, or an increased risk of malignancy. However, as many as 15% of patients with IBM have underlying autoimmune disorders such as Sjögren’s syndrome, SLE, scleroderma, sarcoidosis, variable immunoglobulin deficiency, or thrombocytopenia.^1,3,138

LABORATORY FEATURES

Serum CK is normal or only mildly elevated (usually less than 10-fold above normal).^2,9,126 Positive ANAs and a monoclonal gammopathy of unclear significance are found in approximately 20% of patients with IBM. Antibodies directed against cytosolic 5′-nucleotidase 1 A (cN1A) have been detected in as many as two-thirds of IBM patients, whereas this antibody is very uncommon in other neuromuscular disorders.^139,140 Therefore, cN1A antibody testing may well be useful as a screening test to complement the clinical examination and muscle biopsy, particularly when not all the characteristic features of IBM are present clinically or on muscle biopsy. There is a significant incidence of the HLA DR3 phenotype (*0301/0302) in IBM.¹⁴¹ Skeletal muscle MRI scans demonstrate atrophy and signal abnormalities in affected muscle groups (Fig. 33-10).^75,142 Video-swallow studies in individuals with dysphagia often demonstrate prominence of the cricopharyngeal muscle (Fig. 33-11).

ELECTROPHYSIOLOGICAL STUDIES

Nerve conduction studies reveal evidence of a mild axonal sensory neuropathy in up to 30% of patients.² EMG demonstrates increased spontaneous and insertional activity, small polyphasic MUAPs, and early recruitment.^80,126 In addition, large polyphasic MUAPs can also be demonstrated in one-third of patients, which has led to the misinterpretation of a neurogenic process and misdiagnosis in some patients as having ALS.^126,143,144 However, large polyphasic MUAPs can also be seen in myopathies and probably reflects the chronicity of the disease process rather than a neurogenic etiology.

HISTOPATHOLOGY

Muscle biopsy characteristically reveals endomysial inflammation, small groups of atrophic fibers, eosinophilic cytoplasmic inclusions, and muscle fibers with one or more rimmed vacuoles lined with granular material (Fig. 33-12).^{2,9,126,145–151} Amyloid deposition in vacuolated muscle fibers and to a lesser extent within nuclei can be demonstrated on Congo-red staining using polarized light or fluorescence techniques (Fig. 33-13).^145,146 The number of vacuolated and amyloid-positive fibers may increase with time in individual patients.¹⁴⁷ An increased number of ragged red fibers and COX-negative fibers are also evident in patients with IBM compared to patients with DM and PM and age-matched controls (Fig. 33-12).¹⁴⁸ The myonuclei also appear strikingly abnormal. Some are enlarged, contain eosinophilic inclusions, or are located within the vacuoles and appear to be exploding into the vacuoles themselves.