Neurological manifestations are a well-known feature of coronavirus 2019 (COVID-19).1–4 The peripheral and the central nervous system can be affected. Guidon and Amato provided an excellent update of neuromuscular disorders in COVID-19.5 Guillain–Barré syndrome may be the best characterized peripheral nervous system association at the time of this writing. Dalakas recently reviewed 11 patients with Guillain–Barré syndrome from 4 COVID-19 hotspots, and he also commented on his concerns about myositis from COVID-19.6
The consensus is that the muscle may be affected in patients with COVID-19, but there are few examples of well-characterized patients with COVID-19 and myopathy. Retrospective reports from China disclosed that myalgias or arthralgias occurred in about 15% of patients. Elevations in creatine kinase (CK) occurred in about 14% and were more common in those with severe illness.7 Patients with features of muscle injury had higher neutrophil counts, lower lymphocytes, higher C-reactive protein levels, and higher D-dimer levels.8
Case reports provide additional details about rhabdomyolysis with COVID-19. Jin and Tong reported a 60-year-old man in Wuhan who presented with fever and cough and was found to have ground glass pulmonary infiltrates from COVID-19.9 Laboratory studies included a normal CK on admission. On hospital day 9, the patient noted pain and weakness in the lower limbs and was found to have a CK of 11,842 U/L (normal 38–174 U/L). His urine had occult blood. Pulmonary infiltrates had worsened. He was aggressively hydrated and was continued on treatment with the antiretroviral drug opinavir (lopinavir) and moxifloxacin. Methylprednisolone was added. CK levels started to decline by day 11 and were down to 251 U/L by day 20, consistent with rhabdomyolysis. Suwanwongse and Shabarek10 reported an 88-year-old man from New York who presented to the emergency department with acute onset of pain in both thighs and weakness. He was on simvastatin. He had dry cough and low-grade fever as well as a small unchanged pleural effusion. Laboratory studies disclosed a CK of 13,581 U/L (normal range 120–250 U/L) and positive urine dipstick for blood without red cells. Testing for influenza was negative, and he was found to have COVID-19. He received fluids and intravenous immunoglobulin, and his CK level declined to 368 U/L by day 6.
These patients did not have electrodiagnostic (EDx) testing or muscle biopsies. It is unclear if rhabdomyolysis related to COVID-19 is a form of self-limited viral myositis, a muscle reaction to a systemic immune response, a treatment effect, a form of critical illness myopathy, or some other etiology, but it does seem to be self-limited at least in patients reported. Regarding a possible treatment effect, in a trial of lopinavir and ritonavir in COVID-19, there was no increase in CK in the treatment arm.11 In one of the case reports mentioned above, rhabdomyolysis was present at onset discounting critical illness myopathy. Mao et al8 suggests that muscle injury could be due to the presence of angiotensin-converting enzyme 2 receptors for COVID-19 in skeletal muscle, but there are no data on whether the virus can invade skeletal muscle. Dalakas points out that myopathy from COVID-19 may fall within the spectrum of necrotizing autoimmune myopathy.6 Some patients with COVID-19 are getting corticosteroids and sometimes intravenous immunoglobulin, so they might be treated for autoimmune myopathy. So far, however, the course does not seem consistent with a chronic dysimmune process, but the mechanism of rhabdomyolysis remains undetermined.
Intensive care unit patients with COVID-19 are certainly likely to develop critical illness myopathy, neuropathy, or both.5 The incidence of this occurrence is unknown. At our hospital, I am aware of one patient who underwent EDx testing for severe persistent weakness in the setting of single organ (lung) failure from COVID-19 and after pharmacologic paralysis for 6 weeks (ICU MDs, Philip Lamberty, MD, email communication, May 2020). He also received corticosteroids and aminoglycosides. Not surprisingly, EDx testing showed evidence of sensorimotor axonal polyneuropathy and myopathy believed to be related to critical illness. Such patients are also at risk for prolonged neuromuscular junction blockade. It is unlikely that we will know the incidence of acquired intensive care unit weakness with abnormal EDx studies in COVID-19 patients before there is reluctance to perform EDx testing in that setting.
Many of the COVID-19 patients who suffer from a lengthy illness are also likely to be debilitated from disuse.5 During the COVID-19 pandemic, patients with chronic neuromuscular diseases may also be susceptible to increased weakness because of a reduction to physical activity because of quarantine. With this topic in mind, Stefano et al12 studied 268 quarantined subjects including 149 with neuromuscular diseases and 119 healthy controls. They assessed total weekly physical activity measured as energy expenditure using an adapted version of the International Physical Questionnaire short form, and they assessed physical and mental health by a Short-Form Health Survey (SF-12). Questionnaires were administered by telephone. The International Physical Questionnaire short form assessed frequency and intensity of physical activities ranging from sitting to vigorous exercise both before and during quarantine. Eighty percent of the neuromuscular disease patients were able to walk independently. Forty-six percent had an acquired or hereditary polyneuropathy, whereas 33% had a neuromuscular junction disorder, and 13% had myopathy. The others had degenerative conditions.
As expected, there was a significant decrease in total weekly physical activity in quarantined neuromuscular patients and in healthy controls. In exception, the amount of vigorous activity did not decline in neuromuscular patients, presumably because there were already at a lower level of baseline activity. In neuromuscular patients, physical and mental health measures were reduced in the SF-12, especially in patients with impaired walking. Physical scores were significantly lower in patients with high body mass indices. This study reminds us that although quarantine is protective for our vulnerable patients and reduces spread of COVID-19, it has a negative impact on the mental and physical health and quality of life of those same individuals. The long-term results of quarantine are yet to be evaluated.
Immune-mediated necrotizing myopathy (IMNM) associated with autoantibodies to 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase (HMGCR) is associated with high levels of serum CK and usually proximal muscle weakness. Affected patients are often, but not always, exposed to statins. The spectrum of this IMNM is becoming clearer, but there is still considerable interest in identifying the best treatments. Meyer et al13 reported a retrospective analysis of 55 patients with statin-induced anti-HMGCR IMNM. These patients were seen at several centers in Montreal. Evaluations were extensive, and the authors used univariate logistic regression models to quantify the association of monotherapy maintenance with age, sex, CK levels, strength examination, presence of dysphagia, delay in treatment, and the use of corticosteroids and intravenous immunoglobulin (IVIG) at induction. They then used multivariate logistic regression to identify independent predictors of monotherapy maintenance.
Eighty-four percent of patients received atorvastatin. The median age at diagnosis was 67.7 years. A relatively large proportion—22 patients or 40%—presented with normal strength and elevated CK levels that persisted after statin discontinuation. At the time of treatment initiation, 46 (84%) had proximal weakness. The median CK was 5000 U/L (range 554–23,000). Most patients had histopathologic evidence of necrotizing myopathy. The authors evaluated corticosteroid-based and non–corticosteroid-based induction strategies with the latter including steroid sparing immunosuppressants (SSIs), especially methotrexate and azathioprine, and IVIG. The reader is referred to the article for the extensive analyses of the subgroups and the highlights follow.13
The study is limited by the retrospective design and the lack of a standardized treatment approach as well as the smaller numbers in the individual groups. However, several important points are evident. Many patients with statin-induced anti-HMGCR IMNM had hyperCKemia and normal strength. In such patients who were not treated, proximal weakness often occurred with an associated median delay in treatment of 21.6 months. The data suggest that delay in initiation of treatment decreased the likelihood of successful steroid sparing agent monotherapy maintenance.
When treated, patients were generally induced with either corticosteroids alone, SSI and IVIG, or steroid/IVIG/SSI triple induction. Some patients achieved remission with a corticosteroid-free induction, but the outcomes somewhat favor the use of a triple induction strategy.13 A randomized controlled study will be necessary to define the best treatment for this patient population.
Myopathy, especially necrotizing myopathy, is also one of the immune-related adverse events caused by immune checkpoint inhibitors (ICIs). Myopathy occurs in less than 1% of patients treated with ICIs, and it can be quite severe.14 Dermatomyositis rarely occurs in this setting. As a recent example, Kosche et al15 reported a 63-year-old man with metastatic melanoma who developed back pain and upper and lower extremity weakness 1 day after his first nivolumab infusion. He was given a dose of oral prednisone and was stable. He received a second dose of nivolumab 4 weeks later and developed similar pain along with a diffuse erythematous rash on the trunk and upper extremities. There were papules on the metacarpophalangeal joints, and there were violaceous patches on the upper eyelids consistent with dermatomyositis. CK and aldolase levels were elevated; electrodiagnostic testing showed a length-dependent axonal polyneuropathy and myopathy. Muscle biopsy findings were those of a nonspecific inflammatory myopathy. He improved with steroids and IVIG.
Cytokine release syndrome is associated with chimeric antigen receptor engineered T-cell therapy. A newly reported phenomenon is that dermatomyositis can accompany fulminant cytokine release syndrome after treatment with nivolumab and ipilimumab. Ohira et al16 reported a 70-year-old man with metastatic renal cell carcinoma and new erythema of the hands, extremities, and trunk 1 day after administration of the ICIs. After the second doses, he developed progressive weakness in the arms and legs. Gottron sign, shawl sign, and proximal weakness were found on examination. CK was 17,386 U/L, and the patient also had anti-Mi-2 and antitranscriptional intermediary factor 1-gamma antibodies. He then developed hypotension, fever, highly elevated C-reactive protein, acute renal failure, and features of disseminated intravascular coagulation in the absence of infection—all believed to be consistent with cytokine storm. He received hemodiafiltration for the acute renal failure and improved with the addition of mycophenolate, methylprednisolone pulse therapy, IVIG, and plasma exchange. Although cytokine release syndrome is more commonly associated with chimeric antigen receptor engineered T-cell therapy, clinicians should be aware that it can also occur with the administration of ICIs and in association with dermatomyositis.15
We should also be aware that severe abdominal complications can occur in patients with juvenile dermatomyositis (JDM). Besnard et al performed a retrospective study of 110 JDM patients seen at Necker Hospital in Paris from January 2005 to December 2018.17 Nine (8.3%) presented with 19 different severe gastrointestinal manifestations. A severe manifestation was defined as one that is either potentially life threatening, requires surgery, or requires parenteral nutrition. Four patients had acute pancreatitis, 4 had hepatitis, and 7 had digestive tract disturbances. In 6 patients, these manifestations occurred at the time of JDM diagnosis. All but 2 had active JDM, and all had muscle pain. Regarding the digestive tract involvement, 4 had bowel perforations. All pancreatitis and digestive tract manifestations were believed to be a certain JDM-related event. By contrast, a duodenal perforation that occurred 3 days after high dose corticosteroids may have been treatment related. Hepatitis was generally believed to be related to JDM and not infectious; 1 patient may have had treatment-related hepatitis.
It has been reported that anti-NXP2 autoantibodies are associated with more severe JDM, especially with gastrointestinal involvement.18 In the patients reported by Besnard et al, most had myositis specific antibodies, mainly against NXP2 and antitranscriptional intermediary factor 1-gamma. The underlying pathophysiology of the gastrointestinal events, including pancreatitis, is likely to be vasculopathy, but high doses of corticosteroids might be contributory.17 This report emphasizes that early aggressive management is required in JDM patients who develop severe or sustained abdominal pain.
Two studies detail interesting associations with inclusion body myositis (IBM). Rietveld et al19 described 5 patients with spinocerebellar ataxia (SCA) types 3 and 6 and IBM. Patients developed gait ataxia between ages 18 and 62 years. At 52–65 years, they developed features found to be due to IBM, especially weakness in finger flexors and thighs. Typical muscle histopathologic changes of IBM were seen in 4 patients. Hereditary inclusion body myopathy was essentially excluded by genetic testing. Although the association of IBM and SCA could be due to chance, it is conceivable that IBM may have been triggered by the formation of abnormal polyglutamine strands in these CAG-repeat expansion SCAs. The mechanism by which this would happen in muscle is unclear, however.
Dieudonne et al20 reported a series of patients with IBM and granulomas in muscle biopsy specimens. Investigators from 8 French and Belgian centers studied cohorts that included (1) 23 patients with myositis and granuloma, (2) 23 patients with myositis without identified granuloma, and (3) 20 patients with sporadic IBM (sIBM) without granuloma. Most patients with granuloma myositis had systemic features suggestive of sarcoidosis. Surprisingly, nearly half of them had clinical features of IBM, such as deep finger flexor weakness, and histopathologic features of IBM in addition to noncaseating granulomas. A minority had antibodies against cytosolic 5′-nucleotidase. Of the 10 patients with features of IBM and granuloma myositis, only 1 (10%) responded to immunomodulation, whereas 77% of those with granuloma without IBM features did so. In comparison, 1 of 6 (16%) sIBM patients without granulomas responded to therapy. It is unclear if the extramuscular disease manifestations responded to immunomodulation in the group with granulomas.
The study did not provide any reasons linking IBM with granuloma. It was presumed that patients had sarcoidosis and IBM, and the presence of IBM made the muscle disease unresponsive to immunomodulation. The authors conclude that patients with granuloma myositis should be carefully screened for IBM to tailor specific therapies. One wonders whether the chronic inflammatory process related to granuloma myositis somehow triggered IBM or transformed myositis to IBM.
MUSCULAR DYSTROPHY: FSHD AND SCAPULOTHORACIC SURGERY; ANOCTAMINOPATHY AND HYPERCKEMIA
Patients with facioscapulohumeral dystrophy (FSHD) are sometimes referred for surgical scapular stabilization. Several different procedures have been used. Eren et al21 performed a retrospective analysis of 40 patients (64 shoulders) who had FSHD and underwent scapulothoracic arthrodesis using 5–6 multifilament cables that attached the scapula to the second to seventh ribs and bone graphs between the medial scapula and ribs. Surgical indications were functional deltoid muscles and medial scapular winging, shoulder elevation of <90 degrees, and shoulder elevation of <120 degrees for patients who underwent contralateral scapular surgery (to achieve symmetry).
Outcome assessments involved measurement of shoulder range of motion and preoperative and postoperative forced vital capacities. The mean duration of follow-up was 71 months. Sixty-one shoulders in 38 patients derived significant functional benefit from surgery with a statistically significant improvement in shoulder elevation and in shoulder abduction with the range increasing from 52.7 ± 15.8 to 98 ± 20.3 degrees. Pulmonary function testing, performed in 19 patients, showed no change after surgery.
The postoperative major complication rate was 18.8% (12 of 64 surgeries) with 7 patients requiring reoperation and 5 requiring a chest tube for a mean of 4 days. Seven of the patients had pulmonary complications including pneumothoraces, plural effusions, and atelectasis. Ten had scapular complications including rib fracture with failed scapular fixation. Patients with failed scapular fixation were successfully treated through additional procedures. In addition, 1 patient had a complete brachial plexus palsy after surgery with subsequent recovery of median and radial nerve function after 18 months. There was only partial recovery of ulnar nerve function.
There were no measures of patient satisfaction or change in the quality of life. Nevertheless, this single center study from Istanbul provides level IV evidence that scapulothoracic arthrodesis with the use of multifilament cables is a successful technique in patients with FSHD and scapular instability. Postoperative complications are common and occasionally serious, but generally resolve either spontaneously or with additional procedures. One of 64 operations resulted in brachial plexus.
Mutations in ANO5 present with a limb-girdle dystrophy phenotype or Miyoshi-like muscular dystrophy, but they may also produce isolated hyperCKemia or elevations in CK associated only with exercise intolerance or myalgia. It is unclear if patients with isolated or minimally symptomatic hyperCKemia progress to a muscular dystrophy phenotype over time. To address this unknown, Panades-de Oliveria et al22 retrospectively evaluated 20 adult patients with asymptomatic or mildly symptomatic hyperCKemia and pathogenic mutations in ANO5. Patients were located from the neuromuscular disorders unit database of 2 tertiary hospitals in Spain. These centers followed a total of 27 patients with ANO5 mutations. Interestingly, 20 of the 27 were evaluated because of asymptomatic or mildly symptomatic hyperCKemia. Seventeen of those patients underwent EDx testing and skeletal muscle magnetic resonance imaging (MRI) with 5 having sequential MRIs after 3–5 years. Most patients underwent muscle biopsy.
The mean age at first detection of hyperCKemia was 39 years with a range of 13–66 years. There was a delay in diagnosis with the mean age of diagnosis of anoctaminopathy being 48 years (range 27–69). In some patients, it was known that they had elevations in liver transaminases but not in serum CK. None had weakness, and all had normal neuromuscular examinations. One-quarter had generalized myalgias and another quarter had exercise-induced myalgias. Two had calf pain with calf hypertrophy, whereas 6 had calf hypertrophy without pain. Next generation sequencing of ANO5 revealed 12 different mutations with 7 not being previously reported. CK levels ranged between 250 and 15,000 U/L. Only 6 of 17 patients had EDx studies interpreted as being “myopathic.” In about a third of patients, muscle biopsies showed nonspecific myopathic findings. Three of 19 showed inflammatory changes. Two had mitochondrial or dystrophic features. The imaging results included asymmetric and preferential involvement of the thigh adductors with or without gastrocnemius muscle involvement. Nineteen of 20 showed these typical changes at some point. The abnormalities were seen on the T1 sequences in 3 quarters, but the short tau inversion recovery sequences were more sensitive. Three of the 5 who had serial studies had progressive changes on MRI, and 1 patient who had a previously normal MRI showed new findings of edema in the medial gastrocnemius on follow-up short tau inversion recovery sequences.
HyperCKemia with or without myalgia is commonly encountered in clinical practice. There are numerous potential causes and one is anoctaminopathy. The report by Panades-de Oliveira et al emphasizes that calf hypertrophy can be the only relevant finding on neurologic examination. There is a wide range in CK levels. Age at presentation is highly variable and includes older adults. Electrodiagnostic testing may not be particularly informative, and sometimes muscle histopathology shows inflammatory changes. Skeletal muscle MRI can be helpful in identifying a pattern of weakness involving the thigh adductors and gastrocnemius muscles. Importantly, the study shows that at least the imaging abnormalities progress over time. No genotype–phenotype correlation was found.
1. Whittaker A, Anson M, Harky A. Neurological manifestations of covid-19
: a systematic review and current update. Acta Neurol Scand. 2020;142:14–22.
2. Koralnik I, Tyler KL. COVID-19
: a global threat to the nervous system. Ann Neurol. 2020;88:1–11.
3. Ghannam M, Alshaer Q, Al-Chalabi M, et al. Neurological involvement of coronavirus disease 2019: a systematic review. J Neurol. 2020;19:1–19.
4. Ellul MA, Benjamin L, Singh B, et al. Neurological associations of covid-19
. Lancet Neurol. 2020;19:767–783.
5. Guidon A, Amato A. COVID-19
and neuromuscular disorders. Neurology.
6. Dalakas M. Guillain-Barre syndrome: the first documented COVID-19
triggered autoimmune neurologic disease. More to come with myositis is the offing. Neurol Neuroimmunol Neuroinflamm. 2020;7:e781.
7. Guan W, Ni Z, Hu Y, et al. Clinical characteristics of coronavirus disease 2019 in China. N Engl J Med. 2020;382:1708–1720.
8. Mao L, Jin H, Wang M, et al. Neurologic manifestations of hospitalized patients with coronavirus disease 2019 in Wuhan, China. JAMA Neurol. 2020;77:683–690.
9. Jin M, Tonh Q. Rhabdomyolysis as potential late complication associated with COVID-19
. Emerg Infect Dis. 2020;26:1618–1620.
10. Suwanwongse K, Shabarek N. Rhabdomyolysis as a presentation of 2019 novel coronavirus disease. Cureus. 2020;12:e7561.
11. Cao B, Wang Y, Wen D, et al. A trial of lopinavir–ritonavir in adults hospitalized with severe covid-19
. N Engl J Med. 2020;382:1787–1799.
12. Stefano V, Battaglia G, Glustino V, et al. Significant reduction of physical activity in patients with neuromuscular disease during covid-19
pandemic: the long-term consequences of quarantine. J Neurol. 2020;13:1–7.
13. Meyer A, Troyanov Y, Drouin J, et al. Statin-induced anti-HMGCR myopathy: successful therapeutic strategies for corticosteroid-free remission in 55 patients. Arthritis Res Ther. 2020;22:5.
14. Kao J, Liao B, Markovic S, et al. Neurological complications associated with anti-programed death 1 (PD-1) antibodies. JAMA Neurol. 2017;74:1216–1222.
15. Kosche C, Stout M, Sosman J, et al. Dermatomyositis in a patient undergoing nivolumab therapy for metastatic melanoma: a case report and review of the literature. Melanoma Res. 2020;30:313–316.
16. Ohira J, Kawamoto M, Sugino Y, et al. A case report of fulminant cytokine release syndrome complicated by dermatomyositis after combination therapy with immune checkpoint inhibitors. Medicine.
17. Besnard C, Gitiaux C, Girard M, et al. Severe abdominal manifestations in juvenile dermatomyositis. J Pediatr Gastroenterol Nutr. 2020;70:247–251.
18. Aouizerate J, De Antonio M, Bader-Meunier B, et al. Muscle ischaemia associated with NXP2 autoantibodies: a severe subtype of juvenile dermatomyositis. Rheumatology (Oxford). 2018;57:873–879.
19. Rietveld A, van Gaalen J, Saris C, et al. Inclusion body myositis in patients with spinocerebellar ataxia types 3 and 6. J Neurol Neurosurg Psychiatry. 2020;91:876–878.
20. Dieudonne Y, Allenbach Y, Benveniste O, et al. Granulomatosis- associated myositis high prevalence of sporadic inclusion body myositis. Neurology.
21. Eren I, Ersen A, Birsel O, et al. Functional outcomes and complications following scapulothoracic arthrodesis in patients with facioscapulohumeral dystrophy. J Bone Joint Surg Am. 2020;102:237–244.
22. Panades-de Oliveira L, Bermejo-Guerrero L, Pablo de Fuenmayor-Fernandez de la Hoz C, et al. Persistent asymptomatic or mild symptomatic hyperCKemia due to mutations in ANO5: the mildest end of the anoctaminopathies spectrum. J Neurol. 2020;267:2546–2555.