Clinical manifestations

Presentation and course

Identifying the clinicopathologic syndrome is important to narrow the etiologic differential diagnosis and guide treatment and prognosis. Parainfectious viral myositis may present as several distinct entities: diffuse myalgias (often with little or no weakness), segmental myalgias, polymyositis, inclusion body myositis, rhabdomyolysis, and rarely dermatomyositis, granulomatous myositis, or necrotizing myopathy. Elevations of creatine kinase as well as aldolase, lactate dehydrogenase, and the aminotransferases are expected in many cases of myalgias, most cases of polymyositis and dermatomyositis, inclusion body myositis, and all cases of rhabdomyolysis. For most cases of inflammatory myopathy, the electromyogram shows irritability (fibrillations, positive sharp waves) and low-amplitude, short-duration (“myopathic”) motor units.

A syndrome of acute-onset diffuse myalgias may precede, accompany, or follow infection with various viruses. This association is especially strong for coxsackieviruses and influenza viruses but probably also occurs with various other common human viruses. Sometimes the myalgias may be more restricted, at which point “epidemic pleurodynia” or “Bornholm disease” may be applied (109). Diffuse myalgias also occur in many HIV patients treated with zidovudine and in patients with hepatitis C virus treated with interferon-alpha.

A more specifically defined syndrome with segmental myalgias, sometimes termed “benign acute childhood myositis” or “myalgia cruris epidemica,” can follow infection with influenza A or B viruses (69; 01; 20). This disease characteristically occurs about three days (range 0 to 18 days) after the initial manifestations of influenza (ie, fever, headache, cough, rhinorrhea). Benign acute childhood myositis has a strong predilection for school-aged children, though it has been reported in adults including the elderly.

Most commonly, patients with virus-associated myositis will present with diffuse myalgias, although multifocal myalgias, muscle weakness, and rhabdomyolysis are also possible. There is often muscle tenderness to palpation or with movement, and occasionally there is muscle edema. Most commonly, the large muscles in the legs are affected (quadriceps, calf, gluteus muscle groups) (21). The clinical presentation depends on a number of factors, including which muscles are involved, host characteristics, and virus-specific features. The clinical course of viral myositis may be acute, subacute, or chronic. Myositis associated with influenza and enterovirus infection is most often acute or subacute, whereas that associated with retroviruses and hepatitis is often chronic. Secondary complications, such as rhabdomyolysis, are rare but can occur with any viral myositis.

Another form of segmental myositis involving predominantly the forearm muscles was described in nine adult patients with preceding prodromal symptoms suggestive of viral infection. The muscle involvement was proven by MRI T2 intensity of the forearm flexor and extensor muscles. Although no specific viral agent was detected, the febrile prodrome, and spontaneous complete resolution of symptoms support the diagnosis of segmental viral myositis (112).

Influenza virus. In the case of influenza A or B virus, additional nonspecific systemic symptoms are fever, headache, cough, and rhinorrhea; in most patients who experience myalgias, these are diffuse and self-limited. However, in a subset of patients (from 5.5% to 33.9% of children) a specific syndrome sometimes referred to as “benign acute childhood myositis” will develop, which is distinct from the initial myalgias in that symptoms are more severe, can be focal, and occur later in the course of infection (more than 3 days later).

It is more common in school-aged children with a male predominance (2:1), and 75% are associated with influenza B virus. Children may exhibit sudden-onset calf pain and difficulty walking. There may be tenderness, swelling of gastrocnemius muscles, soleus, or other muscles. Calf muscles are usually involved and in two thirds of cases are the only muscle group involved. These features were shown to be prominent in a 2007-2008 outbreak in the pediatric population in Germany of influenza B-associated benign acute childhood myositis (71).

Benign acute childhood myositis must be distinguished from the more common diffuse myalgia syndrome that precedes or co-occurs with the usual influenza symptoms: (1) benign acute childhood myositis onset is later, after onset of usual symptoms; (2) benign acute childhood myositis is more focal (two thirds of cases involve only the calves), and (3) benign acute childhood myositis affects the patient more severely. Up to 3% of cases of influenza-associated myositis are accompanied by rhabdomyolysis; interestingly, this complication is seen more often in girls with influenza A (contrary to the epidemiological tendencies for benign acute childhood myositis).

A retrospective study in Austria of 375 cases of influenza A H1N1 during the 2009 pandemic demonstrated that a greater proportion of children had serum elevations of creatine kinase (almost 50%) whereas this proportion was much lower in adults (approximately 25%) (96). Meanwhile, myalgias were more often experienced in the group with constitutional symptoms (both adult and children) and were also more common in adults compared to children (all symptom groups). Retrospective analysis of emergency department cases of suspected acute viral myositis in Brazil found that symptoms were most often localized to the legs and in particular the calves with elevated muscle enzymes and leukocytosis (14). During the 2009 pandemic, several cases of acute viral myositis with myalgias, weakness, and elevated CK were reported, all of which recovered within a week with supportive care (42).

Creatine kinase elevation and a myopathic EMG may be useful to narrow the differential diagnosis or to confirm clinical suspicions. Musculoskeletal MRI may show T2 and STIR hyperintensity and contrast enhancement (92). In cases that have gone to biopsy, there is degeneration and necrosis with surprisingly little inflammatory infiltrate. The prognosis is excellent as the disease is self-limited over about 3 days (range 1 to 30 days). Neuraminidase inhibitors, which are useful in influenza only within the first 36 hours of symptoms, have not been studied in benign acute childhood myositis and are not likely to be useful because presentation is usually outside this window.

Myocarditis associated with influenza infection usually develops within the first week of infection. Patients may be asymptomatic or have chest pain, shortness of breath, or signs and symptoms of acute congestive heart failure.

Rhabdomyolysis is a rare but potentially fatal syndrome due to the breakdown of skeletal muscle with the release of skeletal muscle cell contents, including proteins and electrolytes, in the systemic circulation. The rapid release of large quantities of potassium, calcium, organic acids, and myoglobin can lead to renal tubular toxicity and acute renal failure as well as cardiac arrhythmias and compartment syndrome. Numerous viruses have been associated with rhabdomyolyses, including influenza, parainfluenza, enteroviruses, adenovirus, SARS-coronavirus, HIV, herpes viruses, parvovirus, dengue virus, and West Nile virus. Of all rhabdomyolyses associated with viral infection, the most common etiology is influenza virus (42%), followed by HIV and enterovirus. However, a review of 300 cases of influenza-associated myositis found that only 3% developed rhabdomyolysis (01).

MRI findings of intramuscular hemorrhage, mostly in proximal muscles like the shoulder, help differentiate patients with rhabdomyolysis from myositis patients in which edema would be found in the absence of hypointense attenuation (52).

Enterovirus. Infection with coxsackievirus, an enterovirus, is associated with the syndrome of epidemic myalgia or epidemic pleurodynia. Epidemic myalgia manifests as paroxysmal sharp pain in the thoracic and upper abdominal muscles and intercostal regions and is associated with localized muscle tenderness and fever. The pain is worse with cough or deep breathing, and there may be an associated headache and sore throat. Correlating with common times of coxsackievirus infection, this is most often seen in the summer and fall months. A study in Yunnan province in China identified 98 cases of enterovirus-associated acute flaccid paralysis. Most cases occurred in summer months in young children; two thirds had fever at onset, one third had myalgias, and lower numbers had diarrhea or neck stiffness or upper respiratory symptoms (110). Group B coxsackievirus can also cause myocarditis and pericarditis.

Necrotizing myositis caused by enterovirus was described in a post-transplant pediatric patient, raising awareness and the recommendation to look for infection with these viruses in post-transplant patients presenting with an acute neuromuscular disorder (116).

A study using a mouse model delineated the impact of enterovirus 71(EV71) on skeletal muscle, showing that severe disease was associated with myositis, muscle calcification, and persistent motor end plate abnormalities (67).

Retroviruses. The human immunodeficiency virus (HIV) could be associated with myositis.

HIV myositis presents as a subacute-onset, slowly progressive, proximal, often symmetrical, muscle weakness of the arms and legs; this closely parallels the presentation of autoimmune myositis. The serum CK can elevate as much as 10 to 15 times normal; however, it may also be normal.

Although rare, this can be the first presenting symptom of HIV. HIV-associated myositis may occur with seroconversion, as the only indication of a chronic silent HIV infection, or in a known HIV-infected patient. HIV patients are also at risk of medication-induced mitochondrial myopathy and associated myalgias.

In addition, myopathy can occur from the use of antiretroviral therapy, such as in nucleoside-related mitochondrial myopathy.

HIV is also associated with subacute proximal limb weakness due to nemaline myopathy, also known as sporadic late-onset nemaline myopathy (29), or even an inclusion body myositis clinical picture (22). There are case reports of dermatomyositis as a rare presentation of HIV seroconversion (95).

In a small case series of 11 patients with HIV, myositis presented initially with proximal and distal weakness and high CK level resembling polymyositis and improved with treatment. Eventually all cases progressed to a clinical picture strongly resembling inclusion body myositis with distal weakness of fingers and wrist flexors, rimmed vacuoles on biopsy, or anti-NT5C1A autoantibodies. Interestingly, none of these patients developed other systemic autoimmune illnesses associated with polymyositis: interstitial lung disease, Raynaud phenomenon, or arthritis (65).

A retrospective study looked at muscle biopsies of 50 HIV patients diagnosed with biopsy-proven HIV associated myopathy. Three main histological patterns were observed: isolated mitochondrial abnormalities, polymyositis (endomysial lymphocytic infiltrate invading non-necrotic muscle fibers), and nonspecific myositis. The isolated mitochondrial abnormalities are well known to be associated with retroviral medications. None of the patients who had a nonspecific myositis pattern on muscle biopsy developed inclusion body myositis-like features and half of the ones showing polymyositis findings on muscle biopsy eventually developed an inclusion body myositis phenotype in 6 years of follow up. Another interesting finding on this muscle biopsy orientated study was that 82% of patients had detectable viral load at first muscle biopsy, compared with only 22% positive at the last follow up, suggesting the role of retroviral therapy (61).

Human T-cell lymphotropic virus type 1 (HTLV-1) can induce an inflammatory myopathy clinically and pathologically indistinguishable from sporadic polymyositis. This presentation may coexist with the more common HTLV-associated myeloneuropathy/tropical spastic paraparesis or may be the only clinical manifestation of HTLV-1 infection. The association of HTLV-1 with polymyositis is highlighted by studies in two areas where HTLV-1 is endemic: (1) in Jamaica, 7% to 18% of healthy individuals are seropositive compared to 85% of polymyositis patients; and (2) in Kagoshima, Japan, 11.6% of the population is seropositive compared to 27.5% of polymyositis patients. Cases of localized axial (paraspinal muscle) myopathy due to HTLV-1 have been reported (75).

Chronic hepatitis C virus inclusion body myositis has been associated with HIV and HTLV-1, although not nearly as often as polymyositis. The first reports demonstrated the histopathological and immunopathological similarities between sporadic and retroviral-associated inclusion body myositis (22).

Inclusion body myositis has been associated with HIV, HTLV-1, and hepatitis C virus, although not nearly as often as polymyositis. Cupler and colleagues reported inclusion body myositis in patients with HIV and one patient with HTLV-1 (22). They developed all the usual clinical and histological features except for high (up to 1000-fold) elevation of serum creatine kinase. At least seven more cases have been reported (90; 89; 66; 31). Matsuura and colleagues identified positive HTLV-1 serology in 11 of 21 sporadic inclusion body myositis patients in an endemic area of Japan, suggesting that HTLV-1 may be associated with either polymyositis or inclusion body myositis (like HIV) (74). Yakushiji and colleagues found incapacitating inclusion body myositis in a hepatitis C virus carrier and obtained significant improvement using interferon-beta therapy at doses higher than those used by the Muscle Study Group in sporadic inclusion body myositis--250 MIU over 10 weeks compared to 144 MIU over 24 weeks (125). There are several other case reports of inclusion body myositis occurring in the setting of hepatitis C virus infection (104).

Dermatomyositis and granulomatous myositis have not convincingly been attributed to any viral or retroviral etiology. In two HIV patients, a facial rash raised suspicion of dermatomyositis, but classic skin lesions and histologic signs of dermatomyositis were absent (47).

Hepatitis viruses. Hepatitis B virus and hepatitis C virus can be associated with a polymyositis but also with polyarthritis or polymyalgia rheumatica. Chronic hepatitis C virus infection has been associated with myositis. Hepatitis C virus PCR was noticed in infiltrating cells. Patients may have myalgias for some time before the inflammatory myopathy is diagnosed. Chronic hepatitis B virus infection has also been associated with myositis in several cases (80; 94; 88; 13). In one case, lamivudine (an antiviral therapy used in hepatitis B) normalized symptoms and electromyography in a steroid-refractory patient (48).

A case of neuromyopathy associated with acute hepatitis B infection has been described (73).

Inclusion body myositis has long been associated with hepatitis C virus, as evidenced in numerous case reports and series. The clinical presentation and course, histopathological findings on muscle biopsy, and prognosis in patients with sporadic inclusion body myositis and HCV-associated inclusion body myositis are virtually indistinguishable. One study quantified the increased prevalence of positive HCV serologies in patients with inclusion body myositis (28%) compared to matched controls with polymyositis (4.5%), whereas both groups had similar prevalence of positive serologies for HTLV-1, HIV, and hepatitis B virus (115).

A case report of hepatitis E virus-induced severe myositis was reported, with flaccid tetraparesis, acute hepatitis, and renal failure (78).

Other viruses. Alphaviruses (including Ross River virus, Barmah Forest virus, Chikungunya virus) are positive-sense single-stranded RNA viruses of the family Togaviridae. Manifestations in humans include arthritis and arthralgia and myalgias.

A new polyomavirus was identified as a cause of a vasculitic myopathy in a transplant patient. Polyomaviruses are small, double-stranded DNA viruses that are widespread. Most often they cause mild respiratory symptoms or are asymptomatic in humans; however, the two most often associated with human disease are the BK and JC viruses. Mishra and colleagues reported a patient who was on immunosuppression and developed fatigue, myalgias, weakness, and loss of visual acuity (81). Serum CK was initially normal but then elevated to a peak of 8000 U/L. EMG showed widespread myopathy, and muscle biopsy of the biceps brachii showed vasculitis, microthrombosis of capillaries, myonecrosis, myositis, and atrophy.

Parvovirus B19 can cause a myositis with associated fever and diffuse rash (slapped cheeks). Parvovirus B19 infection is usually asymptomatic but may cause aplastic anemia, erythema infectiosum (diffuse rash with “slapped cheeks” appearance and fever), hydrops fetalis, or chronic red cell aplasia. A case of myositis (in the setting of fever, rash, and acute B19 infection) in an adult was described (12).

Adenovirus has been associated with myocarditis, myositis, and rhabdomyolysis. Although herpes zoster ophthalmicus is known to cause ophthalmoplegia, varicella zoster virus and orbital myositis causing ophthalmoplegia in an adult has been reported (58).

Dengue infections occur globally, and symptoms can range from asymptomatic to severe with multiorgan impairment and bleeding. Classic symptoms include rapid onset fever, headache, retro-orbital pain, arthralgia, and severe myalgia. Both direct viral infection and indirect host immune responses are important for the neurologic manifestations of dengue infection. Myalgia is very common, but myositis and weakness are rare and usually severe, accompanied by respiratory muscle involvement, elevated CK, and myopathic findings on EMG.

Dengue-induced myositis has a less aggressive form in the pediatric population compared to adults and should be included in the differential diagnosis of acute flaccid paralysis in dengue-endemic areas (113).

Chikungunya was first identified in 1952 in Tanzania and named by the Makonde people for the painful arthritis that it causes (chikungunya roughly translates as “the disease that bends up the joints”) (100). It is re-emerging in Africa, Asia, South America, and the Caribbean, and more recent outbreaks have been more clinically severe and fatal, due in part to new virus mutations. Chikungunya virus causes fever, rash, arthralgias, and myositis. Symptoms are usually self-limited and resolve within 3 to 4 days.

A case of chronic active Epstein-Barr virus infection causing infiltration of the skeletal muscle was described in a 19-year-old woman with swelling of the trapezius muscle and elevated creatine kinase level. Muscle biopsy of the brachialis muscle showed presence of Epstein-Barr encoded RNA-positive CD8 T lymphocytes (59).

Chronic, active Epstein Barr–related generalized myositis with systemic muscle involvement as the main manifestation is rare, reported in case reports, has a very poor prognosis, and lacks specific treatment. Some cases have been associated with T-cell lymphoma or extranodal natural killer/T-cell lymphoma (106).

Few cases have been reported of myositis caused by cytomegalovirus (CMV) in nonimmunocompromised as well as immunocompromised patients. Acute myositis is an uncommon manifestation of cytomegalovirus, and direct infection of muscles and autoimmunity are likely the pathogenic mechanisms (43; 102).

There were a few case reports of myositis induced by human parechovirus type 3 (HPeV3) in adults and children in Japan. Parechoviruses are members of the genus Parechovirus in the family of picornaviridae and at least 16 types are identified. This virus was associated with myalgia/myositis among adults and children through 2008 to 2017 in different regions of Japan (82).

An unusual case of hantavirus-associated myositis was reported in a 56-year-old man with no medical problems or history of alcoholism, complicated with myocarditis (54). He improved after a high dose of steroids, ciclosporin, and cyclophosphamide.

Coronavirus disease 2019 (COVID-19). Novel coronavirus disease 2019, or COVID-19, is an emerging viral infection that originated in Wuhan, China. The virus quickly spread across the globe within a few months, with significant morbidity and mortality. It is caused by a new coronavirus 2 (SARS-CoV-2). SARS-CoV-2 belongs to the betacoronavirus family, which includes severe acute respiratory syndrome coronavirus (SARS-CoV-1) and Middle East respiratory syndrome coronavirus (MERS-CoV).

SARS-CoV-2, which is responsible for COVID-19 infection; acute respiratory syndrome (SARS-CoV-1); and Middle East respiratory syndrome (MERS-CoV) are members of the betacoronavirus family. Taxonomically they form part of the Coronavirinae subfamily. The Coronavirinae subfamily consists of four genera, and betacoronaviruses include the most pathogenic coronaviruses known to man: SARS-CoV, MERS-CoV, and SARS-CoV-2. They are large, enveloped, positive-sense RNA viruses. These viruses infect humans as well as several groups of animal species.

The transmission of SARS-CoV and MERS-CoV has been attributed to market civets and dromedary camels. SARS-CoV-2 apparently emerged from the wet animal market in Wuhan. All three diseases are believed to originate from bats, although this has been difficult to prove (87). They cause general upper and lower respiratory symptoms as well as gastrointestinal and neurologic manifestations. Human coronaviruses (HCov), which cause human infections, were first discovered in 1965.

Skeletal muscle injury has been reported in all coronavirus infections. Major risk factors for myopathy are severe respiratory distress, systemic inflammatory response, and sepsis. The direct invasion of the muscle by the virus is another potential mechanism for myopathy. Similar to SARS-CoV-1, SARS CoV-2 has the ability to penetrate the cells that express ACE2 receptors. ACE2 is expressed in the muscle cells, making the invasion of the muscles by the virus a possible explanation for myositis and rhabdomyolysis. In addition, the inflammatory changes and cytokine storms in advanced cases of COVID-19 cause immune-mediated muscle damage (105).

Typical symptoms include fever, cough, dyspnea, fatigue, and myalgia.

As many as 60% of patients with SARS CoV infections have myalgia, and up to 30% present with muscle weakness and elevated CPK. There is no statistical difference in CPK levels and severity of lung pathology in patients with SARS CoV infections. Muscle weakness is typically symmetric and involves the proximal muscles as well as trunk and neck muscles. Myopathy and hyperCkemia are frequently reported complications of COVID-19.

A report of 214 COVID-19 patients from Wuhan, China, found evidence of skeletal muscle injury, defined as muscle pain with creatine kinase levels greater than 200 U/L, in 10.7% of patients. In patients with severe infection, the incidence of skeletal muscle injury increased to 19.3% (72). In another group of 95 patients with COVID-19 in Wuhan, the reported incidence of hyperCkemia was 29.5% (130). Patients with muscle injury were noted to have higher neutrophil counts, lower lymphocytes count, higher c-reactive protein, and higher D dimer levels.

In January 2020, ACE2 was identified as the functional receptor for SARS-CoV-2. ACE2 is present in multiple human organs, including the skeletal muscles. However, SARS-CoV-2 using the same receptor was not detected in skeletal muscle in postmortem examination. Therefore, there is a need for additional studies to clarify whether SARS-CoV-2 infects skeletal muscles cells by binding with ACE2. Significantly elevated levels of proinflammatory cytokines may be the culprit for skeletal muscle damage.

There are multiple case reports of rhabdomyolysis associated with COVID-19 infection as well as multiple cases of critical illness myopathy in severe cases of COVID-19 infection (97; 130). Myositis related to SARS-CoV-2 and documented by muscle MRI and muscle biopsy findings has been described in a few cases (10; 129).

Rhabdomyolysis was reported to have a bimodal distribution in patients less than 20 years of age and between the ages of 40 and 60 (49). Two fatal cases were reported. In general, it was seen in patients with a known genetic predisposition or risk factors, such as obesity, hypertension, and diabetes; 77% were male. The median value for CPK was 15,783 IU/L. The incidence of rhabdomyolysis may be between 0.2% and 2.2% of hospitalized COVID-19 patients. Some antibiotics and statins were found to play a role in higher CPK levels.

Creatine kinase elevation can occur in COVID-19 infection in the absence of fulminant rhabdomyolysis and appears to be a marker for poor prognosis.

Few cases of classic dermatomyositis induced by COVID-19 infection have been reported, with skin changes and muscle weakness invariably present. Severe bulbar weakness was reported (129). Myositis specific autoantibodies can be an important clue to diagnosis. The outcomes were variable from milder disease, regaining muscle strength in a week, to severe with full muscle strength regained after 3 months, and some fatal cases associated with significant lung involvement (44).

Anti-MDA5 antibody and anti-NXP2 antibody were more frequently reported in SARS-CoV-2-associated myositis. Other myositis-specific antibodies noticed in COVID-19 patients with myositis were anti-Mi-2, RNP/TIF1γ, and anti-Jo1 (51).

In the middle of a pandemic infection, it is not unusual to draw the conclusion that idiopathic inflammatory myopathies are caused by the infection. There are few case reports of idiopathic inflammatory myopathies associated with COVID-19 infection. There were observations of similar hyperinflammatory state, elevated cytokines, and ferritin as well as macrophage activation syndrome in idiopathic inflammatory myopathies and COVID-19 infection (60).

Orbital myositis was reported as a clinical manifestation of COVID-19 infection (02).

Paraspinal myositis has been reported in patients with COVID-19 infection who underwent MR imaging of the spine for back pain, lower extremity weakness, and lower extremity paresthesia. In all evaluated patients, the MR imaging findings of myositis occurred exclusively in the lumbar spine and involved multiple vertebral body levels. The possible etiology includes direct muscular viral infection with SARS-CoV-2, an immune-mediated parainfectious inflammatory response, and drug-mediated effects versus critical illness myopathy (77).

Prognosis and complications

Prognosis for viral myositis depends to some degree on which virus is involved and the temporal course of the myositis.

Influenza- and Coxsackievirus-associated myositis are usually self-limited, and full recovery is expected. Potential complications include the development of rhabdomyolysis (as described in clinical features above) and the ensuing associated issues. Symptoms often last a week but may last up to a month. Myositis associated with HTLV-1 infection seems to be clinically indistinguishable from sporadic polymyositis and would, therefore, follow a similar natural history.

Myositis occurring in the course of a chronic infection such as HIV, HTLV-1, hepatitis C virus, or hepatitis B virus is usually chronic and will increase the patient’s disability. Early diagnosis and effective management may prevent further disability. Nevertheless, there is concern that using immunosuppression may enhance the severity of the underlying disease and cause more systemic complications or, in the case of AIDS, co-infections. There are a few anecdotal reports that patients with HIV-associated myopathy may improve spontaneously, but it is not known whether this is related to better nutrition, medical care, or other types of supportive therapy. In general, inclusion body myositis associated with HIV, HTLV-1, and hepatitis C virus is like the sporadic form and follows a similar clinical course, with limited response to immunotherapy.

Rhabdomyolysis, myositis, and myalgia in COVID-19 infection have reportedly responded to treatment consisting of hydration, steroids, hydroxychloroquine, azithromycin, and tocilizumab (08; 40; 45; 46). Lower extremity compartment syndrome was reported secondary to COVID-19 myositis in a young, vaccinated woman (11).

A case report revealed the link between creatine kinase, inflammatory markers, and clinical response in a severe COVID-19 infection treated with a combination of intravenous immunoglobulin and baricitinib (124).

Post-acute infection syndrome was reported in patients after COVID-19 infection, and muscular symptoms were present in one study in up to 56% of patients at 6 months.

Prior to SARS-CoV-2, many other viruses had been described to cause post-acute or chronic muscular symptoms. A subset of patients suffering from muscular symptoms had an underlying rheumatological and immunological condition that was latent, undiagnosed, or was sparked off by the acute inflammation during COVID-19 (07).

Clinical vignette

A 32-year-old man presented with several days of proximal leg and arm muscle weakness, calf cramping, and myalgias. He reported low-grade fevers at home but no cough, rhinorrhea, nausea, diarrhea, or rash. Neurologic examination was notable for mild proximal weakness (deltoids 4+/5 and hip flexors 4+/5), but otherwise strength was full throughout. Reflexes were 2+ and symmetrical throughout, and sensation was intact to all modalities.

Laboratory findings were notable for elevated creatine phosphokinase at 1120 units/L (upper limit of normal 225 units/L), aldolase 17.7 units/L (upper limit of normal 8.1 units/L), and positive IgM and IgG for the Epstein Barr virus. Sedimentation rate was normal and C-reactive protein was elevated at 0.72mg/dL (upper limit of normal 0.5mg/dL). A myositis serology panel was negative (including Mi-2, PL-12, PL-7, EJ, OJ, SRP, Ku, U2 snRNP, PM/SCL, Jo-1). Other findings included: low-level cryoglobulinemia, polyclonal immunoglobulin pattern on serum protein electrophoresis, normal white blood cell count, mildly positive ANA at 1:160, normal TSH, and serologies for E Chaffeensis negative (IGG, IGM), RMSF/Rickettsia, C. pneumoniae IGG positive, IGM negative, HHV6 IGG positive, IGM negative, Babesia duncani IGG negative, parvovirus B19 IGG positive, IGM negative, and Lyme western blot negative.

Electrodiagnostic studies revealed normal motor and sensory nerve conduction studies, and on needle electromyography, there was an early recruitment of low amplitude and short duration motor unit potentials without abnormal spontaneous activity, such as positive sharp waves or fibrillations.

Based on clinical history examination and laboratory and EMG findings, he was diagnosed with Epstein Barr-associated myositis. He was treated symptomatically, and weakness resolved within days; myalgias and cramps resolved within 6 weeks. Creatine phosphokinase returned to normal. Repeat electrodiagnostic study after 3 months was normal. Muscle biopsy was not performed due to full recovery.

Biological basis

Etiology and pathogenesis

Human viruses associated with inflammatory myopathy include coxsackieviruses, influenza A and B viruses, HIV, HTLV-1, hepatitis C virus, and hepatitis B virus. Rare cases have been associated with cytomegalovirus (70), Epstein-Barr virus (114), and West Nile virus (107). The specific clinicopathological syndrome may narrow the likelihood of a viral etiology.

Viral contributions to inflammatory myopathy are postulated to occur via several mechanisms. The mechanisms of viral myositis are not fully elucidated, but there is evidence for several pathophysiological mechanisms. The virus-host interactions that contribute to myositis include: (1) direct infection (acute or chronic) or host response to viral antigens, (2) molecular mimicry, and (3) immune dysregulation. This section covers the host and pathogen contributors to the pathogenesis of viral myositis.

Direct infection. In most of the studies published to date, there was not enough evidence of direct viral invasion of the muscle fibers.

In inoculated mice, Coxsackie viruses can cause acute and chronic myositis (103). Furthermore, muscle cultures (especially immature myotubes) can be directly infected with coxsackieviruses. Nevertheless, the role of coxsackieviruses in human polymyositis has been unconvincing, with some reports of high antibody titers in chronic inflammatory myopathies (16). Similarly, there has not been any clear evidence that influenza virus causes sporadic polymyositis in humans although in two well-studied cases, viral particles were seen in muscle biopsy by electron microscopy (57). Maeda and colleagues reported a case of myositis occurring in the setting of acute cytomegalovirus infection (70). In two cases, hepatitis C virus RNA was detected in muscle biopsies using polymerase chain reaction (118). Mumps virus was once considered a candidate for causing inclusion body myositis because the 15-nm to 20-nm microtubular filaments in inclusion body myositis resemble mumps virus nucleocapsids (15), but the search for mumps virus via polymerase chain reaction has been negative (62; 39). The hepatitis C virus RNA and antigen were detected in the inflammatory cells attacking muscle fibers in a case of hepatitis C virus–associated inflammatory myopathy (34). Leff and colleagues also thoroughly searched for nucleic acids of encephalomyocarditis virus (see below), adenovirus, HIV, HTLV-1, and HTLV-2 in 44 polymyositis patients; all were negative (62). Electron microscopy, immunocytochemistry, in situ hybridization, cultures, and polymerase chain reaction of muscle biopsy in patients with inflammatory myopathy and HIV, HTLV-1, or hepatitis B virus infection have failed to reveal evidence of virus (64; 48). However, in two cases of inflammatory myopathy associated with chronic hepatitis B infection, hepatitis B virus DNA and viral antigens were found inside intact muscle fibers (13). HTLV-1 proviral DNA was detected in CD4+ T cells around, but not in, myofibers in a case of inclusion body myositis (74). Although HIV, HTLV-1, and hepatitis C virus do not seem to cause persistent muscle infection, they do persist in many cells of the body and may cause muscle inflammation via several mechanisms: (1) immune cell activation, leading to cytokine and lymphokine release that may induce the expression of nontolerant antigens on muscle cells or (2) molecular mimicry. During the COVID-19 pandemic, there was evidence that some COVID-19-infected patients develop multifactorial myalgia and weakness and even elevated CK suggestive of an inflammatory myopathy similar to HIV-associated cases (25). There has not yet been a convincing clinicopathological series of inflammatory myopathy in COVID-19-infected patients except in rare case reports.

Molecular mimicry. Molecular mimicry was first studied as a possible mechanism of inflammatory myopathy when antibodies to the Jo-1 autoantigen were discovered in up to 10% of patients with myositis (24). Jo-1 autoantigen, part of a histidyl-transfer RNA synthetase (111), has structural homology with the genomic RNA of encephalomyocarditis virus, an animal picornavirus related to coxsackieviruses and enteroviruses. This association was strengthened when enteroviral RNA was found by in situ hybridization in muscle fibers from patients with myositis (127). However, polymerase chain reaction studies have repeatedly failed to confirm the presence of enterovirus in muscle biopsies of inflammatory myopathy (62; 63). Nevertheless, a French study using reverse transcriptase PCR found enterovirus RNA in three of 20 muscle biopsies of inflammatory myopathy patients, compared with 0 of 29 controls (35). In HIV, gag and pol genes share antigens with human ribonucleoproteins, which are the target of circulating autoantibodies in some sporadic polymyositis cases (101).

Immune dysregulation. Retroviruses including HIV and HTLV-1 as well as hepatitis viruses might act via immune dysregulation. There are clinical and histological similarities between HIV or HTLV-1 myopathy and sporadic polymyositis (37; 37), leading Dalakas and colleagues to perform comparative immunohistological studies in patients with HIV myositis, HTLV-1 myositis, and sporadic polymyositis (64). The predominant inflammatory cells within the endomysial infiltrates were CD8+ T cells (49% of the total cells) and macrophages (38%), which surrounded or invaded non-necrotic muscle fibers. In four patients with inclusion body myositis and HIV, it was shown that a percentage of the autoinvasive CD8+ T cells showed clonal restriction specific for HIV virus antigens (31). These findings suggest that chronic infection may trigger a T cell-mediated inflammatory process (128), and studies have begun uncovering the genetic mechanisms by which this might occur with coxsackieviruses in mice (103).

Virus-host interactions. Influenza viruses are the most common infectious agents responsible for viral myositis. Influenza viruses are members of the Orthomyxoviridae family and are single-stranded negative-sense RNA viruses. The actual pathogenesis of the symptoms of influenza-associated myositis is unknown but may be in part related to elevated levels of circulating cytokines. Transient viremia allows access to muscle cells by the virus, and there may be a predilection for the calf muscles in “benign acute childhood myositis.” Muscle biopsy shows degeneration or rhabdomyolysis with or without inflammation, which is focal and patchy. Mouse models have shown infection of myocytes with influenza virus. There are variable reports of isolating influenza virus and detecting antigens from human muscle biopsy samples. That the myositis more often occurs in children is postulated to be because young muscle is more permissive to infection. Influenza B is more often associated with myositis, and that may be due to certain myotropic glycoproteins in its structure. Myocarditis from influenza may be due to the direct effects of virus on the myocardium, or exacerbation of underlying coronary artery disease. Rhabdomyolysis in the setting of influenza infection is proposed to be due to several possible factors: direct viral invasion of myocytes, cytokine storm from the immune response, or toxicity of circulating viral factors. Viral isolates from the 2009 influenza A pandemic compared to a seasonal influenza A (both H1N1) were studied. One group found that that the influenza A virus can directly infect human muscle cells in vitro, and in this model, levels of inflammatory cytokines were not increased after infection (32). In addition, this group shows that the pandemic virus replicated at higher titers than the seasonal virus and had a lytic effect on the muscle cells. Interestingly, the authors also detected on the surface of human primary muscle cells, sialic acid receptors, which are the same as those on respiratory epithelium and which serve as the receptors to which the influenza virus binds.

Coxsackieviruses A and B are enteroviruses, which are single-stranded positive-sense RNA viruses in the Pircornaviridae family and include polioviruses and echoviruses. Enteroviruses are transmitted via the fecal-oral route and include an ever-enlarging number of serotypes. Most epidemic pleurodynia is caused by group B Coxsackieviruses and only infrequently by group A 4, 6, 9, 10, or the enteric cytopathogenic human orphan (ECHO 1, 6, 9, 16, 19) viruses. A study in Yunnan province in China assessed 98 cases of enterovirus-associated acute flaccid paralysis; the researchers identified 33 serotypes of enterovirus and characterized their phylogenetic relationship to the prototype and molecular evolution compared to strains in other parts of the world (110). In inoculated mice, Coxsackieviruses can cause acute and chronic myositis.

Polymerase chain reaction studies in muscle biopsy specimens have been inconclusive for confirming the presence of enteroviruses (both positive and negative results) and attempts to isolate retroviruses have been negative.

Alphaviruses (including Ross River virus, Barmah Forest virus, Chikungunya virus) are positive-sense single-stranded RNA viruses of the family Togaviridae. In a mouse model, the Ross River virus causes an upregulation of inflammatory cytokines and causes severe myositis, whereas Barmah Forest virus did not cause this upregulation and is not associated with replication in myocytes (50).

A study showed that interleukin 17 (IL-17) expression was increased in musculoskeletal tissues and serum of Ross River virus–infected mice and humans (85). The blockade of IL-17 using a monoclonal antibody could reduce the disease severity.

In vitro studies of muscle cells infected with Chikungunya virus show upregulation of many proteins involved in metabolism, cell signaling, and notably the cytoskeleton (53). Mouse models infected with Chikungunya virus show development of edematous muscles, viral antigen detected in muscle connective tissue fibroblasts and satellite cells (91), myoblasts and muscle fibroblasts (19), and upregulation of proinflammatory cytokines that correlates with severity of disease (86; 33). An elegant study of the pathogenic mechanisms of the epidemic strains from the La Reunion outbreak in 2006 was analyzed compared to a strain from 1983 outbreak in Senegal, and both were able to infect muscle connective tissue fibroblasts, yet only the La Reunion strain directly infected the myofibers with high viral titers in muscle leading to myonecrosis. The reason for this difference is unclear as tropism was similar, and induced cytokine and chemokine profiles were also similar (99).

Polyomaviruses are small double-stranded DNA viruses that are widespread. Most often they cause mild respiratory symptoms or are asymptomatic in humans; however, the two most often associated with human disease are the BK and JC viruses. A novel polyomavirus has been identified in endothelial cells causing vasculitic myopathy and retinal blindness in a pancreatic transplant recipient. A double-stranded DNA virus was isolated from muscle tissue with greatest homology to chimpanzee polyomaviruses (tentatively named New Jersey polyomavirus 2013) (81).

Retroviruses are single-stranded positive-sense RNA viruses that replicate by reverse transcription of the RNA genome into a double-stranded DNA molecule that integrates into the host genome. HIV and HTLV are human retroviruses that have been associated with inclusion body myositis. Muscle biopsy of HIV patients with inclusion body myositis show similar pathology to idiopathic inclusion body myositis with endomysial infiltration with CD8 cytotoxic T cells and macrophages, which invade non-necrotic muscle fibers that are MHC Class I antigen expressing. Through a variety of molecular and biochemical techniques, viral antigens were not detected in the muscle fibers and only at times in the surrounding macrophages. Instead, in HIV- and HTLV-associated inclusion body myositis, the underlying pathophysiology is thought to be driven by clonal-activated T-cells that attack MHC-class I expressing myofibers. Myotoxicity may be enhanced by increased secretion of cytokines and chemokines (26). Clonal expansion of viral-specific T-cells surrounding muscle fibers was found in patients with HIV-associated inclusion body myositis. A significant proportion of the autoinvasive CD8+ cells invaded myofibers expressing a particular human leukocyte antigen-A allele (31; 27). In HIV, gag and pol genes share antigens with human ribonucleoproteins, which are the target of circulating autoantibodies in some sporadic polymyositis cases.

There are four HTLV-1 subtypes, A-D according to geographical origin. Most persons are asymptomatic lifelong carriers, but HTLV-1 can cause severe adult T-cell leukemia or HTLV-1 associated myelopathy (tropical spastic paraparesis) and is associated with several inflammatory disorders, including myopathy.

Dengue virus is a member of the single-stranded RNA Flaviviridae family, transmitted by mosquitos of the Aedes genus. Muscle biopsy has not shown myocyte invasion by dengue virus, with lymphocytic infiltrates at foci of myonecrosis (117).

Toscana virus (Bunyaviridae family) is a Mediterranean arbovirus transmitted by sandflies. Infected patients are usually asymptomatic or present with fever and myalgia. Two cases of myositis and fasciitis have been reported (84).

Genetics. There are no known definite genetic predispositions to viral myositis in humans. In a mouse model of viral myositis, innate immune pathways are being investigated. SHP-1 normally suppresses macrophage-mediated inflammation and contributes to muscle disease. In SHP-1 knock-out mice, although there is still myofiber infection and inflammation with the TMEV virus, the mice do not develop myonecrosis and retain ambulation. This correlated with immature macrophages in the SHP-/- mice compared to the mature macrophages in the wild-type mice. This suggests that SHP-1 promotes inflammation through maturation of macrophages and, thus, is necessary for virus-induced myonecrosis with clinical consequences (120).

Patients with underlying hereditary myopathies or muscular dystrophies are at higher risk of more severe myositis during viral infection. Patients with dysferlinopathies may be especially at risk for Coxsackie B virus infection (119). In certain cases, a viral infection and myositis may be the first presenting sign of a hereditary myopathy. Appropriate cases should be screened when clinically indicated.

Epidemiology

Benign acute childhood myositis is overwhelmingly a pediatric disease, though there are reports of adults including the elderly being affected as well (126). Some studies suggest benign acute childhood myositis may occur in 6% to 34% of childhood cases of influenza A or B. It appears to be more common with influenza B and has a 2:1 male predominance. Up to 3% of cases of influenza-associated myositis are accompanied by rhabdomyolysis; interestingly, this complication is seen more often in girls with influenza A (contrary to the epidemiological tendencies for benign acute childhood myositis).

The factors that predispose a patient with hepatitis C virus, HIV, or HTLV-1 infection to develop a myopathy are unknown. The use of zidovudine in HIV has increased the incidence of myopathy in HIV patients, but this myopathy is different from HIV-associated myositis. The frequency of HTLV-1-related myopathy in the United States is unknown but may be expected to gradually increase with globalization and exposure to endemic areas (Caribbean, Africa, Japan). HTLV-1 may be endemic to the southern United States (20).

Prevention

There are no specific measures to prevent the development of an inflammatory myopathy once viral infection has occurred.

Primary prevention of the many viruses discussed is the only sure method to prevent viral myositis.

Whether or not neuraminidase inhibitors can prevent the development of benign acute childhood myositis following influenza A or B exposure or infection is unstudied.

Clinicians should be mindful of postvaccination adverse events, varying from mild to severe.

Differential diagnosis

Confusing conditions

Viral-associated myositis should be distinguished from other acquired myopathies occurring in noninfected populations. Metabolic, hereditary, endocrine, or toxic causes should be excluded. Systemic symptoms (such as fever, upper respiratory symptoms, or gastrointestinal symptoms) will often distinguish viral myositis from other causes. Myotoxic factors, such as medications, illicit drug, other co-infections, and the effects of multiple organ failure that occur in HIV-positive patients with AIDS should also be considered. Interferon may play a role in some patients with hepatitis C virus infection (76; 04).

Other causes of infectious myositis (bacterial, fungal, or parasitic) should also be considered. Although rhabdomyolysis is a rare complication of viral myositis, the differential diagnosis for rhabdomyolysis includes trauma, prolonged immobilization, excessive muscle activity, compartment syndrome, heat exposure, drugs of abuse (cocaine, heroin, amphetamine), alcohol, medications (statins, fibrates, salicylates, steroids), and idiopathic inflammatory myopathies.

Associated or underlying disorders

Underlying hereditary myopathy may predispose persons to viral myositis and in the appropriate clinical setting (ie, glioma) should be excluded.

Myopathy with the features of a mitochondrial disease at biopsy can occur in the setting of zidovudine or didanosine therapy for HIV; in fact, this is the most common myopathy in HIV patients (79). This is not a parainfectious process, per se, but deserves mention as it would feature prominently in the differential diagnosis of any HIV patient with neuromuscular weakness. The clinical presentation is indistinguishable from HIV-associated myositis. The distinction is made via muscle biopsy showing ragged-red fibers, subsarcolemmal red-rimmed cracks or pale granular washed-out staining, lipid accumulation, and focal depletion of cytochrome c oxidase (COX); and electron microscopy showing accumulation of abnormal mitochondria with paracrystalline inclusions and excessive amounts of lipid droplets (79). This probably results from depletion of mitochondrial DNA (mtDNA), as shown by genetic analyses (05). Notably, up to 8% of all HIV patients on zidovudine will develop diffuse myalgias. Whether this is a prodrome to zidovudine-associated mitochondrial myopathy is not known. However, antiretroviral drugs (nucleoside analog reverse transcriptase inhibitor, NRTI) used to treat human HIV patients, lead to mitochondrial myopathy, with COX-negative fibers and with evidence of multiple mtDNA deletions (93). In both cases, it seems clear that the effects of NRTI drugs damage the maintenance of mtDNA proximity.

Pyomyositis bacteria myositis should also be considered in AIDS and immunocompromised patients. This focal infection is usually due to Staphylococcus aureus or rarely to Gram-negative organisms (122). Pyomyositis is suspected when patients have low-grade fever, localized pain, and swelling in a large muscle group; leukocytosis and creatine kinase elevation may or may not occur. Imaging with ultrasound, MRI, or CT will show some combination of edema, fluid density, and contrast enhancement. Risk factors include underlying muscle abnormalities, local trauma, exercise, hematogenous spread of a systemic bacterial infection, carpal tunnel syndrome, or a seizure type or finding.

Other factors that might contribute to the development of a myopathy in AIDS (and possibly chronic hepatitis B and C) patients include vitamin deficiencies, disuse atrophy, chronic wasting syndrome, and myocytotoxicity caused by septicemia, bacteriotoxins, and antimicrobial drugs, ie, hyperlipidemia.

Diagnostic workup

As with any other acquired form of inflammatory myopathy, the laboratory tests that complement the clinical examination in viral myositis include the following:

	(1) Determination of serum muscle enzymes (CK, AST, ALT, LDH, aldolase)
	(2) Electromyography. Electromyography in an inflammatory myopathy usually shows spontaneous activity (fibrillation potentials, positive sharp waves) and myopathic motor unit potentials (short-duration, low-amplitude, polyphasic), and an early recruitment pattern
	(3) Muscle biopsy. Muscle biopsy is not needed in all cases of suspected viral myositis. If a self-limited and acute viral infection with associated myositis is suspected (such as with influenza or coxsackievirus infection), it is advisable to wait a few weeks before considering muscle biopsy as symptoms may completely resolve. Muscle biopsy is especially important in HIV as it is the critical test to distinguish HIV-associated myositis from medication-induced myopathy. Diagnostic evaluation must be tailored to the individual presentation and history.

In the case of specific viruses, additional diagnostic considerations are in order.

Influenza infection is often diagnosed clinically, based on activity in the community, but also by polymerase chain reaction (PCR) of nasal swabs. Muscle enzymes are often elevated (CK, LDH, AST). EMG shows myopathic changes, and in cases where muscle biopsy was performed, there is degeneration and necrosis with minimal inflammatory infiltrates. In cases with myocardial involvement, the EKG may show nonspecific tachycardia, ST elevation, Q waves, or left bundle branch block. Transthoracic echocardiogram may show left ventricular dysfunction, wall motion abnormalities, and reduced ejection fraction. Cardiac MRI may show focal left ventricular edema and contrast enhancement. Myocardial biopsy at this stage will show foci of active inflammation, edema, degeneration, and necrosis, and influenza virus can be detected by PCR.

Coxsackievirus infection can be assessed by serological testing and culture of both pharyngeal and fecal specimens. Although it is not generally recommended for diagnosis, muscle biopsy may reveal detection of viral antigens.

In the case of a hepatitis (B or C) patient who has symptoms and signs of myositis, gamma-glutamyl transpeptidase (GGT) should be checked in addition to the transaminases, as GGT is specific for liver injury whereas AST and ALT may be elevated in muscle or liver diseases. Conversely, a polymyositis patient with GGT elevation should be investigated for a chronic hepatitis virus infection.

Muscle biopsy in myositis associated with HIV and HTLV-1 is often indistinguishable from sporadic polymyositis. Rarely, a few cytoplasmic bodies and rods can be seen. Occasionally, myofiber degeneration predominates and endomysial inflammation is sparse. Scattered angular fibers, if present, suggest the co-occurrence of an axonal neuropathy. Vacuoles have been described in HTLV-1-associated myositis (123).

Muscle is often affected in patients with HIV or AIDS. In a prospective clinicopathological study, 60% of untreated HIV-positive patients without neuromuscular symptoms, and seemingly with normal strength, had histological abnormalities in muscle, such as mild inflammation, type II fiber atrophy, or denervation (41). In another study of 50 untreated patients without neurologic deficits but with some degree of muscle wasting, 96% had histological abnormalities consisting of denervation (76%), type II atrophy (58%), endomysial inflammation (36%), and necrosis with phagocytosis (30%). These changes are probably multifactorial in origin and may be the histological correlates of the myalgia, fatigue, diminished endurance, or transient CK elevation noted in AIDS patients without weakness. Likely patients with this disease are underevaluated, and the true burden of disease is unrecognized.

A retrospective study looked at muscle biopsies of 50 HIV patients diagnosed with biopsy-proven HIV associated myopathy. Three main histological patterns were observed: isolated mitochondrial myopathy, polymyositis, and nonspecific myositis. The isolated mitochondrial myopathy was associated with retroviral medications (61).

Suspected myositis in a patient with HIV will often require muscle biopsy. Serum elevation of muscle enzymes and myopathic pattern on EMG support the diagnosis. Muscle biopsy will show inflammatory infiltrates of T-cells and macrophages in the endomysium, with or without necrosis. Biopsy can help distinguish HIV-associated myositis from other myopathies associated with HIV, such as that from ART (nucleoside-related mitochondrial myopathy), nemaline myopathy, and inclusion body myositis.

COVID-19-associated myositis and rhabdomyolysis may be confirmed by nucleic acid amplification test or validated antibody testing (36). Currently, no SARS-CoV-2 viral particles have been reported in muscle biopsies.

Microscopic examination of muscle in patients who died from COVID-19 showed type 2 fiber atrophy in 32 out of 35 patients, necrotizing myopathy in nine, and myositis in seven. The myositis was defined by perivascular and endomysial inflammatory cell infiltrates. In patients with myositis CD68 positive, CD4-positive or CD8-positive histiocytes and T cells were observed more frequently than CD20-positive B cells. Diffuse MHC-1 immunostaining of nonnecrotic/nonregenerating muscle fibers was evident in all 16 patients with myositis or necrotizing myopathy and in eight additional patients. One patient exhibited MHC-1 staining predominantly in perifascicular fibers, finding seen in dermatomyositis; however, there was no abnormal MxA expression or clinical findings of dermatomyositis. SARS-Cov2 nucleocapsid immunohistochemistry was negative in all 35 cases (108). These findings suggest as prior documented that muscle damage in SARS-Cov2 infections is secondary to an inflammatory response, including cytokine damage.

Another case control autopsy study showed that patients who died from COVID-19 infections had signs of myositis on a spectrum ranging from mild to severe inflammation. The upregulation of MHC class I antigens in the early phase of the disease and the concomitant upregulation of MHC class II antigens on myofibers in later stages indicate involvement of skeletal muscle in the immune response against SARS-CoV-2. Some patients showed capillary expression of MxA, indicating a type I interferon signature and a perifascicular expression of MHC antigens similar with dermatomyositis findings (06).

Management

In the case of acute viral myositis, most treatment is symptomatic and supportive. Patients with mild myopathic symptoms and signs that are not functionally limiting can be monitored with serial examinations and CK levels. Many of these patients remain stable and do not require specific therapy. Nonsteroidal anti-inflammatory drugs may be useful in managing myalgias.

Patients with HIV-associated myositis may benefit from HIV-specific antiviral therapy, including zidovudine and other medications that may cause a distinct myopathy. However, there is one report of zidovudine worsening an HIV-associated myopathy (09). Treatment with antivirals for other viruses, such as Epstein Barr virus, should be considered on a case-by-case basis and treatment individualized.

With the chronic myositis associated with HIV, patients may respond to immunotherapy. Treatment with prednisone (as in sporadic polymyositis) may be considered and does not seem to worsen HIV or immunocompromised status (55). Nevertheless, intravenous immunoglobulin is an attractive alternative because it is well tolerated and immunoenhancing rather than immunosuppressive. The largest case series suggests that 50% of patients will achieve remission after nine months (98). Other immunotherapies, such as IVIG and immunosuppressants, have been evaluated in HIV-associated myositis, but there are no definitive data, and there is concern over further immunosuppression in patients with HIV-associated myositis.

Zidovudine-associated myopathy (a mitochondrial myopathy distinct from HIV-myositis) will often improve if the dose is reduced, with or without a drug holiday. Improvement generally begins within one to two months and may be incomplete, even if the offending drug is stopped permanently. A wide array of effective antiretroviral alternatives is now available to the HIV-positive patient.

HTLV-1-associated polymyositis is treated the same as sporadic autoimmune polymyositis, with less concern for immunosuppression complications than with HIV infection. Myositis does not appear to be due to direct, persistent retroviral infection of myofibers but, rather, an inflammatory reaction induced by the HTLV-1 infected monocytes. Two patients with HTLV-1 associated myositis were treated with alemtuzumab, a monoclonal antibody against the CD52 molecule on B and T cells, monocytes, macrophages, and eosinophils. HTLV-1-associated polymyositis is typically refractory to multiple immunotherapies. Two patients with refractory HTLV-1-associated polymyositis were treated with alemtuzumab and showed clinically meaningful improvement in muscle strength, reduction in CK to normal, and normalization of myopathic EMG findings (17).

Special considerations

Pregnancy

Though HIV-associated myositis could certainly occur in a pregnant woman, to our knowledge there are no reported cases of this or other virus-associated myopathies in this population.