Article In Brief
Postmortem studies of people who died after severe COVID-19 did not show signs of the virus, but did show extensive immune-mediated and inflammatory damage in muscles and nerves.
People who died following a severe case of COVID-19 have extensive inflammatory and immune-mediated damage in their psoas muscle and femoral nerve, but no signs of the SARS-CoV-2 virus, a study published online June 7 in Neurology found.
The Harvard-led study of 35 consecutive autopsies of patients who died following COVID-19 in April to June of 2020 compared their histopathology findings to those of 10 controls who died in the intensive care unit of other causes.
Only four of the patients had neuromuscular symptoms before death, but in most cases their COVID-19 infections were so severe there was no opportunity for a neuromuscular exam.
On autopsy, laboratory values were available in 27 of the patients. Peak creatine kinase was elevated in 20 patients (mean 959 U/L, range 29-8413 U/L). Major histocompatibility complex-1 (MHC-1) expression was seen 16 cases of necrotizing myopathy or myositis and eight additional patients, for a total of 24 out of 35.
In addition, abnormal expression of myxovirus resistance protein A (MxA), known to be induced due to viral infection, was present on muscle capillaries in nine patients and on nerves in seven.
The main question left unanswered by the study is how the results compare to those in patients who survived COVID-19, particularly when treated with the steroid dexamethasone.
“These were patients who died in the ICU during the first wave of COVID-19 in Boston,” said the senior author of the paper, Isaac H. Solomon, MD, PhD, assistant professor of pathology at Harvard Medical School. “These were the worst-case scenarios. We need confirmatory studies in patients who survive. Follow-up studies of patients who survived acute COVID 19 and died of other causes will tell us whether these abnormalities persist, particularly in the long-haulers.”
Neurologists familiar with the paper said it's the largest neuromuscular pathology study in COVID-19 patients to date, and they welcomed the findings.
“Up until this point, we've seen only small case reports with few patients,” said Dianna Quan, MD, FAAN, professor of neurology and director of the electromyography laboratory at the University of Colorado School of Medicine.
“This took everybody who died of COVID-19 in a window of time from April and June,” Dr. Quan said. “I don't think it proves that the virus doesn't get into the muscle and nerves, but it is suggestive that we're dealing mostly with an immune-mediated attack.”
All 35 patients with SARS-CoV-2 infection who died between April 5, 2020 and June 13, 2020 and who subsequently underwent autopsy at Brigham and Women's Hospital were included in the study. The mean age at death was 67.8 years old, with a range of 43 to 96 years. Symptom onset to death ranged from one to 49 days.
Five of the 35 patients had received remdesivir, as did “four or five” who received hydroxychloroquine; one patient was in a placebo-controlled trial, the paper reported. One patient received dexamethasone, but not specifically for COVID-19.
In the seven patients with myositis, the study found that “CD68-positive, CD4-positive, and/or CD8-positive histiocytes and T cells were observed more frequently than CD20-positive B cells.” Four of the nine patients with necrotizing myopathy had abnormal MxA immunostaining, as did three of seven with myositis, and two without either condition.
The mean time from the onset of COVID-19 symptoms to death was 12.8 days in patients with necrotizing myopathy, 17.1 days in patients with myositis, and 18.1 days in those with neither finding.
Dr. Solomon's group tested for SARS-CoV-2 via nucleocapsid protein immunohistochemistry, but results were negative in all 35 cases, the study found. Microscopic examination of the femoral nerve found neuritis in nine of the 35 patients, four of whom also had myositis.
“Perivascular inflammatory cells were observed in six patients, endoneurial infiltrates in one, and both perivascular and endoneurial inflammatory cells in two,” the paper reported. “CD68-positive histiocytes were most abundantly observed in all cases, but were sometimes co-predominant with CD8-positive, or less often, CD4-positive T cells. MxA immunostaining was observed in 7/35 (20 percent) of cases in the capillaries, only one of which had neuritis. SARS-CoV-2 IHC was negative in all 35 cases.”
Four of the nine patients with neuritis had a history of diabetes, as did two with a history of connective tissue disease and one with neither condition. Two of the 35 patients had a history of polyneuropathy. One of the patients with diabetes had received chemotherapy for acute lymphoblastic leukemia, but had no inflammation of the nerve.
The mean time from onset of COVID-19 symptoms to death was 13.4 days in patients with neuritis, compared with 17.6 days in patients who did not have symptoms.
Although Dr. Solomon and colleagues found no evidence of virus in the nerve or muscle samples, the paper noted that an earlier study of diaphragm muscle found the RNA of SARS-CoV-2 in four of 26 consecutive autopsies. The discrepancy, the paper concluded, “may be explained by differences in methods to detect the virus or examination of different muscles.”
Another paper, published online first in October 2020 in Lancet Neurology, reported SARS-CoV-2 immunostaining in cranial nerves (glossopharyngeal and vagal nerves), but only in two of 40 patients. That finding, the paper noted, raises “the possibility that viral infection of peripheral nerve may occur. In that study, SARS-CoV-2 immunostaining was found in undefined cells within the medulla from which these cranial nerves originate. Contiguous spread of infection from the medulla to these cranial nerves is conceivable.”
It is also possible, the paper noted, that virus had penetrated into muscle or nerve, but that it had cleared by the time of death. However, Dr. Solomon told Neurology Today: “This study included a range of different illness durations prior to death, and we didn't see anything. What we do see is activation of the body's antiviral response.”
The paper concluded: “Although we did not measure cytokine levels in blood, the histopathological abnormalities seen in our patients suggest that these findings may be secondary to the storm of cytokine release rather than direct viral infection of these tissues. Further studies are needed to better understand the pathogenic mechanisms of myopathy and neuropathy associated with SARS-CoV-2.”
Waqar Waheed, MD, professor and vice chair of neurological sciences at the University of Vermont College of Medicine, said the paper's findings of inflammatory cytokines fit well with those reported earlier this year in Cancer Cell. That paper found high levels of inflammatory cytokines and markers of neurodegeneration in the CSF of cancer patients who had been hospitalized for over a month with moderate to severe neurological symptoms associated with COVID-19.
“These two papers had similar findings regarding high levels of inflammatory cytokines and no evidence of virus in either the CSF or peripheral nerve,” Dr. Waheed said.
The new paper, he said, provides potentially useful information for therapeutic intervention. “An exaggerated type 1 interferon response appears to be a factor in severe cases of COVID-19,” Dr. Waheed said.
A complicating factor, however, is that in mild COVID-19 infections, interferon can actually be beneficial, he said. “The key thing we need is to find some biomarkers that can predict progression of COVID-19 disease from mild to severe,” Dr. Waheed said. To that end, future studies should include detailed neuromuscular examination and measures of neurophysiology, which was not in the current paper, he said.
In the meanwhile, Dr. Waheed added, “Studies are testing the effectiveness of recombinant interferon in the early stages of COVID-19 infection, as well as the use of baricitinib, which inhibits the interferon 1 signaling pathway, in patients with severe Covid-19.”
Urvi G. Desai, MD, professor of neurology and director of Carolinas MDA Care Center at Atrium Health in Charlotte, NC, said, “Interestingly, COVID 19 symptoms to death was shorter in the necrotizing myopathy group compared to myositis and/or a group with neither of these findings, while it was four days sooner in neuritis group, which may indicate that necrotizing myopathy and neuritis are poor prognostic markers, if present.” “It's hard to extrapolate to surviving patients, but we need to be careful in patients in whom there is an indication of necrotizing myopathy or neuritis,” Dr. Desai said.
“This is an important attempt to provide insight into how SARS-Cov 2 infection affects the peripheral nervous system. The authors help us to understand if a direct viral invasion of muscle and nerve is a culprit versus a cytokine storm triggered an immunological inflammatory response,” she added.
MxA expression was observed in muscles and nerves in endothelial cells, she noted, suggesting a host response to clear viral RNA. Based on the MxA expression seen in nine muscle and seven nerve samples, an overexpression of interferonopathy could have been a player in inflammatory and necrotic responses.
She also noted that the study did not test for the membrane attack complex, which could have indicated whether or not the complement system was in play
“These were patients who died in the ICU during the first wave of COVID-19 in Boston. These were the worst-case scenarios. We need confirmatory studies in patients who survive. Follow-up studies of patients who survived acute Covid-19 and died of other causes will tell us whether these abnormalities persist, particularly in the long-haulers.”
— DR. ISAAC H. SOLOMON
“It is also going to be important to know the pathophysiology at play in other viral myositis,” Dr. Desai said. “Does it differ, and if so, do these findings have any clinical implications?
Drs. Solomon, Waheed, and had no disclosures. Dr. Quan has received compensation for honoraria/travel related expenses and compensation to her institution for her research for Alnylam, Pfizer, Ionis, Argenx, Momenta, Cytokinetics, and IngenioRx.