Introduction
The coronavirus pandemic began in December 2019 with Wuhan, China, being the epicenter.[[1]] As per the latest reports on August 8, 2020, there have been 20,806,961 cases worldwide
with 747,258 deaths due to COVID-19.[[2]]
Many reports have indicated extrapulmonary involvement of severe acute respiratory
syndrome-coronavirus 2 (SARS-CoV-2).[[3]],[[4]],[[5]],[[6]],[[7]] The common neurological manifestations include headache, anosmia, hypogeusia, and
altered consciousness while seizures, encephalitis, meningitis, encephalopathy, demyelination,
and stroke are potential serious complications.[[8]],[[9]]
Acute disseminated encephalomyelitis (ADEM), an acute inflammatory autoimmune demyelinating
disease of the central nervous system (CNS),[[9]],[[10]] is often preceded by a viral infection, but the evidences depicting its occurrence
as a complication of SARS-CoV-2 are scarce.[[10]],[[11]] Our study aims to provide an overview of this potential but serious complication
of COVID-19.
Methods
A systematic search was conducted for research articles on COVID-19-associated ADEM.
Three primary databases were used, Pub-Med, LitCovid, and WHO. The search strategy
used the keywords, corona virus, COVID-19, neurological complications, and acute disseminated
encephalomyelitis, and was comprehensive with cross-checking of reference lists from
the articles retrieved. Selected articles were independently reviewed by two authors.
All disagreements were resolved with discussion between the two authors and mutually
agreed upon by the authors. The Preferred Reporting Items for Systematic Reviews and
Meta-Analyses guidelines were used [[Figure 1]].[[12]] The final review articles fulfilled the following criteria:
Figure 1: PRISMA diagram
-
Reported ADEM in COVID-19-positive patients
-
Full-text, peer-reviewed articles (case studies and case series).
Articles which did not have patient data such as in vitro studies and articles with
unclear inference were excluded.
This study did not require ethical approval as data were obtained from already available
databases and patients were not directly involved.
Risk-of-bias assessment
Risk-of-bias assessments were performed at the outcome measurement level during data
collection. The National Institute of Health scale was used for case series and case
report studies[[13]] [[Table 1]].
Table 1: Quality assessment of the included studies
Results
A total 135,160 articles were found from the database using the abovementioned keywords
until August 8, 2020. After de-duplication, 73,500 articles were screened and 4932
were found to be pertaining to neurological manifestations of COVID-19. Out of which,
1585 were focusing on COVID-19-associated acute disseminating encephalomyelitis. After
the initial screening, articles were reviewed, out of which 680 articles matched the
inclusion criteria. These articles were then screened in the second review. Before
the qualitative synthesis, 668 articles were removed from the listed articles as they
did not contribute to the manuscript's objective. The final analysis contained 14
articles [[Figure 1]]. Two independent reviewers reviewed all articles. Out of the 14 articles, 11 articles
are case reports, 1 article is a case series, 1 is a retrospective cohort study, and
1 is a webpage/news article.
Novi et al.,[[14]] Parsons et al.,[[15]] Zoghi et al.,[[16]] Zanin et al.,[[17]] and Zhang et al.[[18]] reported cases of ADEM following COVID-19 while Abdi et al.[[19]] reported the ADEM case without prominent clinical pulmonary symptoms. The case
reported by Utukuri et al.[[20]] had no respiratory symptoms of COVID-19 while Reichard et al.[[21]] described the case where ADEM was confirmed by autopsy. Sharma[[22]] reported India's first COVID 19-associated ADEM. McCuddy et al.[[23]] presented a case series of three cases of COVID-19-associated ADEM while Paterson
et al.[[24]] described the retrospective cohort study of neurological manifestations of COVID-19
including ADEM. Yong et al.[[25]] reported the cases of acute hemorrhagic leukoencephalitis while Poyiadji et al.[[26]] described acute hemorrhagic necrotizing encephalopathy [[Table 2]].
Table 2: Study characteristics
Table 2: Contd...
Table 2: Contd...
Table 2: Contd...
Table 2: Contd...
Demographic data
From the 21 total cases, 10 (47%) were male patients and 11 (53%) were female patients.
Female:Male ratio is calculated to be 0.9:1. The age range of the patients was 21–71
years with a mean age of 51.36 years. Diabetes mellitus was the most common comorbidity
seen in four patients, followed by hypertension seen in three patients. Other comorbidities
were ischemic heart disease (1) and dyslipidemia (1). One patient had a past history
of anterior communicating artery aneurysm clipping 20 years ago. Four patients had
no prior illnesses, and details of 10/21 patients were not mentioned in the studies.
History of COVID-19
Out of 21 cases of ADEM, the diagnosis of SARS-CoV-2 was confirmed in 18 and suspected
in 3. Out of 18 confirmed cases, 14 patients had positive oropharyngeal swab SARS-CoV-2
reverse transcriptase–polymerase chain reaction (RT-PCR), 1 patient had positive serum
SARS-CoV-2 RT-PCR, 1 patient had positive cerebrospinal fluid (CSF) SARS-CoV-2 RT-PCR,
and 2 patients tested positive for serum SARS-CoV-2-immunoglobulin G (IgG) with a
negative oropharyngeal swab.
Only 15 patients presented with prior symptoms of SARS-CoV-2 infection. One patient
contracted the infection in postoperative period while they were unspecified in three
patients. Fever was the most common presenting symptom (9), followed by cough (6),
fatigue, and myalgia (3).
Symptoms/clinical presentation
The duration of COVID-19 symptoms ranged from 3 days to 25 days before onset of neurological
symptoms. Among the neurological symptoms, altered consciousness was most common (7),
followed by anosmia (3), paraplegia (3), brain stem involvement (3), sphincter involvement
(2), and quadriplegia (1). Only two patients directly presented with neurological
symptoms.
Laboratory investigations
Blood investigations and CSF analysis were done in 17 patients. Raised inflammatory
markers were most commonly seen in nine (ferritin raised in four, C-reactive protein
in five, and D-dimer in five), and lymphopenia was seen in four patients.
CSF analysis was not reported in four patients. The reasons for the same were traumatic
tap (1), raised intracranial pressure (ICP) (1), pandemic reason (1), and unknown
reason (1).
Eight patients had normal CSF analysis on routine microscopy. Out of nine abnormal
reports, increased CSF protein level was the most commonly reported abnormality in
six, followed by lymphocytic pleocytosis in four patients. Meningitis was ruled out
by negative CSF bacterial cultures in 5 and viral panel in 15 out of 17 patients.
One CSF sample grew Staphylococcus capitis which was probably contaminated.
Although there was neurological involvement in all cases, CSF SARS-CoV-2 PCR was positive
only in 1 patient out of 13 who were tested. One patient had brain tissue CoV-2 PCR
negative on autopsy.
CSF autoimmune analysis was done in 17 patients. Only one patient had a CSF-IgG ratio
of 1.35, one patient had four oligoclonal bands (OCBs), while one patient had mirror
pattern of serum and CSF IgG. Antibodies for myelin oligodendrocyte glycoprotein (MOG)
and aquaporin-4 were negative in nine patients.
Radiological investigations
CNS imaging was abnormal in 19 patients and unavailable in 2 patients. Diffuse white
matter changes, occasionally involving the deep gray matter, were the most common
finding. Out of 21 cases of ADEM spectrum, 13 were ADEM, 7 were AHEM (acute hemorrhagic
encephalomyelitis [AHEM]), while 1 was acute necrotizing hemorrhagic leukoencephalopathy
(ANHLE).
Treatment
Fifteen patients were treated with corticosteroids, 11 patients received IVIg, while
3 patients received convalescent plasma. Two patients needed surgical intervention:
one patient underwent decompressive craniectomy (ADEM with hemorrhage) and the other
one underwent external ventricular drainage.
Complications
Complications included seizures (1), acute kidney injury (AKI) and septicemic shock
(1), raised ICP (1), and supraventricular tachycardia (SVT) secondary to hydroxychloroquine
(HCQ) (1).
Outcome
One patient recovered completely and one had good recovery with mild deficits. Thirteen
patients had incomplete recovery while three patients died as the consequence of the
disease. The cause of death in these three patients was status epilepticus, septicemic
shock with multiorgan failure and severe necrotizing encephalitis respectively.
Discussion
The clinical spectrum of COVID-19 is commonly represented by the respiratory system,
but it also has myriad extrapulmonary manifestations, and neurological involvement
is not unknown. Similar to other coronaviruses, SARS-CoV-2 also has neurotropic potential
and the neurological manifestations of SARS-CoV-2 include anosmia, headache, seizures,
cerebrovascular accidents, meningitis, encephalitis, ADEM, Guillain–Barre syndrome,
and cerebral venous thrombosis.[[27]]
SARS-CoV-2 may directly invade the CNS, or it may cause a parainfectious autoimmune
disease. ADEM is one such parainfectious disease affecting the CNS. It is commonly
seen in pediatric population with few cases reported in adults. The pathogenesis of
ADEM is still unclear. The most likely mechanism is autoimmunity triggered by cross-reaction
of viral antigens to myelin proteins. AHEM or ANHLE are variants of ADEM which have
a fulminant course.[[9]],[[10]]
Recent literature on COVID-19 suggests that it also can have neurological manifestations
with mechanisms similar to other members of its family. Yeh et al. reported the first
case of ADEM associated with coronavirus OC43 in 2003 where the virus could be isolated
from the CSF.[[28]] Arabi et al. reported a case of ADEM following a Middle East respiratory syndrome
infection in 2015.[[29]] The routes of spread to the CNS can be hematogenous or retrograde via the olfactory
nerves. The virus then uses angiotensin-converting enzyme 2 receptors on endothelial
cells to enter the brain.[[5]],[[6]],[[7]],[[30]] In an autopsy study, Paniz-Mondolfi et al. confirmed the presence of SARS-CoV-2
virus particles in the brain. They observed these viral particles in the endothelial
cells of the frontal lobe of the brain.[[31]]
We had 18 confirmed cases of SARS-CoV-2, out of which only one patient had positive
CSF SARS-CoV-2 PCR.
ADEM is a diagnosis of exclusion and needs high level of suspicion. The medical fraternity
is overwhelmed with the current situation of the pandemic, and rare diagnoses may
not seem very obvious unless specifically considered. According to a study by Pohl
et al. and Silvia Tenembaum et al., ADEM is characterized by prodromal symptoms, followed
by acute-onset encephalopathy along with multifocal neurological deficits.[[20]],[[21]],[[32]],[[33]],[[34]] We studied 21 patients of ADEM in SARS-CoV-2 infection and found that 71.4% (15)
patients had a viral prodrome before the onset of neurological symptoms. Fever (9)
was the most common presenting symptom, followed by cough (6), fatigue, and myalgia
(3). These may not always be present in every case of ADEM and some patients may directly
present with the neurological symptoms. Lin et al. studied 50 cases of ADEM and concluded
that previous history of upper respiratory tract infection was seen only in 50% of
the cases.[[35]] In our study, Abdi et al.[[33]] and Utukuri et al.[[20]] described one case each where the patient directly presented with neurological
symptoms while Reichard et al.[[21]] described a case where an elderly patient contracted the infection postoperatively.
Wender et al [[10]] in their study mentioned that the presenting clinical features of ADEM were altered
sensorium, seizures with focal neurological deficits like speech difficulty and limb
weakness. Our findings were parallel to the literature and that altered sensorium
was the most common neurological presentation seen in seven patients. Brain stem involvement
indicates severe disease and was seen in three patients; all were elderly patients
aged 48, 51, and 59 years.[[15]],[[24]],[[36]] Other neurological symptoms in our study were anosmia, limb weakness, and sphincter
involvement. Less common symptoms were limb ataxia described by Utukuri et al.[[20]] in one case and vision impairment described by Novi et al.[[14]] in one case. Zhang et al. in their case report described a case with seventh and
lower cranial nerve involvement with gaze abnormality.[[18]] Gaze abnormality was also seen in another case described by McCuddy et al.[[23]]
Blood investigations were mentioned in 17 patients. We found that the inflammatory
markers were raised in 9/17 patients. It indicates pro-inflammatory state likely from
the preceding infection. Four patients had lymphopenia suggestive of a viral infection.
CSF analysis in suspected cases of ADEM is done to rule out infection or CNS pathologies.
While CSF picture in ADEM is characteristic for its lack of abnormality, it may be
abnormal in almost half of the patients.[[37]],[[38]],[[39]],[[40]],[[41]] The abnormal CSF findings were mild lymphocytic pleocytosis and elevated proteins
which were seen in 30-40% of the patients.[[32]],[[34]],[[42]],[[43]],[[44]],[[45]] In our study, 17 patients underwent CSF analysis which was normal in eight patients
(47%). Out of the nine abnormal CSF analyses, raised CSF proteins were seen in six
patients and lymphocytic pleocytosis was seen in four patients.
Bacterial meningitis was ruled out categorically in 5 out of 17 patients while viral
panel was negative in 15 out of 17 patients. SARS-CoV-2 virus could not be isolated
from the CSF samples. PCR was performed in 13 patients out of 21 and it was positive
only in 1 patient reported by Novi et al.[[14]] Paterson et al. reported one case of ADEM with hemorrhage who needed decompression
craniectomy where they obtained brain tissue for examination. They too could not isolate
the virus in the brain tissue.[[24]]
Only one autopsy study by Paniz-Mondolfi et al. provides evidence for the presence
of SARS-CoV-2 virus particles in the brain. They observed viral particles in the endothelial
cells of the frontal lobe of the brain.[[31]]
CSF and Autoimmune analysis is an important part of investigations in ADEM cases.
OCBs are inconsistently found in CSF in patients of ADEM. They have been reported
only in 10%–50% of cases.[[37]],[[38]],[[39]],[[46]] The presence of OCBs is suggestive of evolution to multiple sclerosis (MS).[[47]] In our study, CSF autoimmune workup was available in 17 patients. Parsons et al.[[15]] described one patient of ADEM who had 4°CBs in CSF.
Mirror pattern of OCB suggests that they are found in CSF as well as the serum. It
is not specific for the diagnosis of ADEM. It only indicates the same autoimmune process
in the serum as well as the CSF. Novi et al.[[14]] described another patient of ADEM with positive mirror pattern of OCB. Among the
serum autoimmune studies to diagnose ADEM, high serum anti-MOG antibodies are found
in almost half of the patients with ADEM, although they too are not specific to ADEM.[[10]],[[48]]
The diagnosis of ADEM is based on multifocal CNS involvement, alteration of sensorium
(encephalopathy), and brain and spine magnetic resonance imaging (MRI) consistent
with demyelinating lesions. Imaging is an important pillar in the diagnosis of ADEM.
The MRI findings of ADEM in the studies by Pohl et al [[32]] and TENEMBAUM et al [[34]] were multiple asymmetric bilateral hyperintensities, involving subcortical and
deep white matter, grey white matter junction, deep grey nuclei, corpus callosum,
brain stem, cerebellum and in one third of the cases even long segments in spinal
cord. Bilateral thalamic involvement is seen in ANHLE which is a fulminant variant
of ADEM. It is rapidly progressive and has high mortality.[[49]]
Brain lesions may be small or large confluent. They are usually hyperintense on T2-weighted
and fluid-attenuated inversion recovery sequences and at times may show minimal enhancement.
We found that almost all patients had bilateral asymmetric extensive lesions involving
the subcortical and deep white matter. Novi et al.[[14]] described one patient who had optic nerve enhancement in postcontrast MRI. It occurs
due to breakdown of the blood–brain barrier. We found that contrast-enhancing lesions
were seen in 7 out of 13 patients of ADEM, 5 out of 7 patients of AHEM, and 1 patient
of ANHLE. Contrast enhancement is thus suggestive of a severe disease as seen in AHEM
and ANHLE. Four patients also had features of myelitis on imaging involving long segments.
MS is a differential diagnosis of ADEM. Although there are no absolute criteria, it
is important to differentiate ADEM from MS. Characteristics Multiple Sclerosis lesions
are periventricular in location, solitary, perpendicular to corpus callosum and 'black
hole appearance' on T1.[[32]] Gray matter involvement and absence of CSF OCB also help in ruling out MS. History
is also important, and MS does not follow a viral illness or vaccine. Other differentials
to consider are MOG antibody-associated disease, neuromyelitis optica spectrum disorder,
Progressive multifocal leukoencephalopathy–John Cunningham virus, and sarcoidosis.[[50]]
To date, there have been no randomized controlled trials and treatment of ADEM is
not standardized. High-dose steroids are the first line of treatment, followed by
intravenous Ig (IVIG) and plasmapheresis in nonresponders. High-dose steroids consist
of intravenous methylprednisolone administered at 30 mg/kg/d (maximum 1 g/day for
5 days), followed by oral taper for 4–6 weeks that has shown to be effective in 60%–80%
of the cases.[[32]],[[34]],[[51]],[[52]]
In our study, eight patients received combination therapy with IV methylprednisolone
as well as IVIg. There was no mortality in any of these patients, and all of them
showed partial recovery. Four patients received only IVIg, while other four patients
received only IV methylprednisolone and two patients received convalescent plasma.
Acyclovir was given in two, remdesivir in one, HCQ in two, and antibiotics in four
patients. Treatment details of one patient were unavailable. Literature shows that
supportive empirical treatment with acyclovir is given in some cases of encephalitis
till infective cause can be ruled out. The same, however, does not apply for empirical
use of antibiotics.[[53]]
We found that four patients developed complications. They were status epilepticus
(1), AKI and septicemic shock (1), raised ICP with mass effect and midline shift (1),
and acute hydrocephalus and SVT (1). Out of these, AKI and septicemic shock were secondary
to SARS-CoV-2 infection in a case described by Reichard et al.[[21]] They describe a 71-year-old male patient who contracted SARS-CoV-2 postoperatively
and subsequently developed ADEM. His course was complicated by respiratory failure
and AKI secondary to septicemic shock. He required continuous renal replacement therapy
but eventually succumbed.
Abdi et al.[[33]] describe a case of a 58-year-old male patient diagnosed as ADEM, who had status
epilepticus and finally succumbed. Paterson et al.[[24]] describe a case with AHLE who developed raised ICP and required decompressive hemicraniectomy.
Another patient of AHLE developed SVT secondary to HCQ treatment which could be managed
medically. The same patient had also intraventricular hemorrhage and obstructive hydrocephalus
for which he needed an external ventricular drain.[[36]]
ADEM has a favorable outcome in 10%–50% of patients who may show complete recovery.[[37]],[[38]],[[40]] We observed that 14% (3) patients had good outcome at the end of treatment while
66.6% (14) patients had partial ongoing recovery including all those with AHLE, some
patients may be left with residual neurological deficits or may even have a relapse.
In case of a relapse, the likelihood of MS should be kept in mind.
The mortality rate of ADEM is reported as 5%–12%[[37]],[[38]] while another study reported it to be 25%.[[54]] Adults have a more serious course of ADEM as compared to children. The rapidly
progressive cases such as ANHLE or AHEM can be fatal. In our study, mortality was
14% (3) while the outcome of one patient remains unknown at the time of writing this
article.
Our study proves to be an important evidence in the literature where data on ADEM
as a rare complication of COVID-19 are scarce. However, these findings could be confirmed
by larger scale studies. Another limitation of our study is inclusion of data from
webpages/news report article[[22]] as the official case reports of few cases were unavailable at present. This could
reflect poor quality of the included study.
