Introduction
Coronavirus disease (COVID-19) is caused by the novel virus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [
1]. Since its recent discovery in Wuhan, China, coronavirus disease has spread across the world, leaving physicians challenged by its variable clinical manifestations.
Most patients infected by SARS-CoV-2 have presented with a mild clinical course: beginning with fever and dry cough, progressing to a form of mild or moderate respiratory disease, and resolving without specific treatment [
2]. Serious complications of the infection, however, remain a central concern. Acute respiratory distress syndrome, acute heart injury or failure, acute kidney injury, sepsis, disseminated intravascular coagulation, and life-threatening metabolic derangements have all been reported in COVID-19 patients, particularly among those with underlying comorbidities or advanced age [
1,
3].
As knowledge of SARS-CoV-2 and its clinical appearance continue to grow, the literature has shown a significant number of infected patients exhibit neurological symptoms [
4,
5]. In this systematic review, we evaluate various neurological manifestations reported in COVID-19 patients and hypothesize their underlying pathophysiology. We deem the timeliness of this systematic review relevant, given the state of the COVID-19 pandemic, but encourage readers to consider the implications of early review and analysis in the clinical setting.
Methods
Our systematic review utilized the PRISMA (Preferred Reporting Items for Systemic review and Meta-Analysis) statement in conjunction with the PRISMA checklist and flow diagram for manuscript format development [
6].
Literature search
The following databases were reviewed for published studies prior to May 20, 2020: PubMed, Google Scholar, and China National Knowledge Infrastructure (CKNI). We also searched pre-print servers including Research square, medRxiv, SSRN, and ChinaXiv. Boolean logic was used for conducting database search and Boolean search operators “AND” and “OR” were used to link search terms. The following search strategy was adopted: COVID-19 OR SARS-CoV-2 OR 2019-nCoV OR nCoV OR novel corona AND neurological OR neurologic OR brain OR CNS OR nervous AND manifestation OR symptoms OR presentation. Titles, abstracts, and full text were screened to ensure they met eligibility criteria. Two authors (GN and JHR) screened, retrieved, and excluded reports. Additional investigators were consulted if uncertainty arose during the review process.
Eligibility criteria
We included any study, published in any language, which reported neurological manifestations in patients infected by SARS-CoV-2. This included case reports and pre-print publications. We excluded all review articles, hypotheses papers, and papers reporting neurological symptoms in MERS-CoV and SARS-CoV patients.
Data was manually extracted from eligible studies by the research investigators. The following variables were included: first author, type of design, site of study, year of publication, published journal or pre-print server, sample size, and reported neurological manifestations.
Outcome measures
Our outcome was to elucidate the neurological manifestations of COVID-19 reported in the medical literature. The results were divided into three categories: central nervous system manifestations (e.g., headache, encephalopathy, and stroke), peripheral nervous system impairment (e.g., dysfunction of taste, dysfunction of smell, neuropathy), and skeletal muscle manifestations (e.g., myalgia).
Discussion
Coronavirus contains four structural proteins, including the spike protein (S), envelope protein (E), membrane protein (M), and the nucleocapsid protein (N). Among them, the S protein plays the most important role in virus attachment, fusion, and entry. SARS-CoV-2, like SARS-CoV, recognizes the angiotensin-converting enzyme 2 (ACE2) as a host cell entry receptor [
41]. High expression of the ACE2 receptor is seen in type II alveolar cells of the lung, intestine, esophagus, cardiomyocytes, proximal tubular cells, and urothelial cells [
42]. Glial cells and neurons have been reported to express ACE2 receptors, making the brain a potential target of COVID-19 infection [
43].
How the virus enters the central nervous system is still a subject of debate. One plausible route of entry is through the olfactory nerve. Retrograde transfer into the axon, whether through synapses, endocytosis, or exocytosis, could explain viral migration into the brain [
44,
45]. Experimental studies using transgenic mice suggest that SARS-CoV and MERS-COV may enter the central nervous system intranasally through the olfactory nerve. Once in the brain, they have the potential to spread to the thalamus and brain stem, two regions highly involved in coronaviridae infections [
46‐
48]. A study evaluating specific receptors in the nasal mucosa, however, suggest SARS-CoV-2 may reach the brain through mechanisms independent of axonal transport via olfactory sensory nerves [
49].
Another theory suggests if SARS-CoV-2 gained access to the general circulation, it could potentially invade the cerebral circulation and continue viral spread [
43]. Slow movement in the cerebral microvasculature may promote interaction of the SARS-CoV-2 S protein and capillary endothelium ACE2 receptor. Once bound, the virus would have the potential to infect, damage, and bud from the capillary endothelium, thereby facilitating viral entry into the cerebrum [
43,
50].
SARS-CoV-2 could also cause damage to the central nervous system indirectly. Viruses do not have to enter the brain to cause damage; they can activate an immune response that triggers subsequent damage within neuronal tissue. SARS-CoV-2 has been reported to cause a massive release of cytokines, a syndrome known as “cytokine storm”—downstream effects of this immune response include endothelial damage, disseminated intravascular coagulation, and disrupted cerebral auto-regulation [
51,
52].
Common symptoms elucidated from our systematic review include myalgia, headache, altered sensorium, hyposmia, and hypogeusia. Acute viral respiratory tract infections have already been shown to cause the aforementioned neurological symptoms [
53]. Peri- and post-infectious hyposmia and hypogeusia is hypothesized to be secondary to olfactory nerve and/or apparatus damage from direct insult of viral infection [
54]. According to the multicenter prospective European study by Lechien et al., it appears that a large number of COVID-19 patients experience olfactory and gustatory symptoms. In their study, olfactory and gustatory dysfunction occurred in 85.6 and 88.8% of patients, respectively, with a majority of patients (65.4%) experiencing abnormalities in olfaction after the appearance of their general symptoms [
36].
We found that in COVID-19 patients, altered sensorium and encephalopathy were not uncommon. The basic pathological change seen in this disease is cerebral edema, with key clinical features being headache, confusion, delirium, loss of consciousness, seizure, and coma. COVID-19 patients with encephalopathy were, by large, older male patients with cardiovascular comorbidities and severe infection with systematic inflammation and multi-organ dysfunction. Early identification of COVID-19 patients with altered sensorium is critical, as underlying potential reversible causes, including impending respiratory failure, require timely intervention [
55]. The pathophysiology behind the cerebral dysfunction is hypothesized to be in part inflammatory-mediated [
56]. This is supported by the fact that the encephalopathic Italian patient had a dramatic response to high-dose steroids [
29].
Ischemic stroke is another clinical entity which can present in patients with COVID-19 infection. This presentation may arise secondary to a cytokine storm syndrome [
51], which can cause endothelial damage, disseminated intravascular coagulation, and disrupted cerebral auto-regulation [
52]. Through the ACE2 receptor of the vascular endothelium, the virus's extensive invasion of the vascular endothelium obviously may cause extensive endotheliitis, increasing the risk of thrombosis leading to ischemic stroke. Critically ill patients with severe SARS-CoV-2 infection often show elevated levels of D-dimer, a fibrin-degraded product which serves as a marker of dysfunctional activation of the coagulation system, such as in acute ischemic stroke [
1,
5,
57]. However, it is observed that the stroke prevalence in COVID-19 patients and non-COVID-19 patients are similar. Two retrospective studies in Wuhan, China, reported prevalence rates of between 2.8–5.0% of acute ischemic stroke [
5,
7] which is consistent with the stroke prevalence in Northeast China prior to the outbreak [
58,
59]. Although the biochemistry suggests an association of COVID-19 with disorders of the coagulation system, further studies are warranted to establish a relationship.
For the three cases of COVID-19-associated intracerebral hemorrhage, a few potential explanations exist. It is known that the expression and the ability of ACE2 receptor to lower blood pressure is reduced in patients with hypertension. In SARS-CoV-2 infection, the presence of S protein could further reduce the expression and function of ACE2 proteins. This could potentially lead to uncontrolled hypertension, arterial wall rupture, and cerebral hemorrhage in infected patients [
56]. Moreover, if the virus disseminates within the cerebral microvasculature, subsequent damage of capillary endothelial cells could result in a tear of the vasculature sufficient enough to cause parenchymal hemorrhage [
43]. Additionally, COVID-19 patients have been reported to have thrombocytopenia and coagulopathy, both of which are contributory factors for secondary brain parenchymal hemorrhage [
5,
60].
It is still unknown how COVID-19 causes encephalo-myelitis. It is thought that once viral particles gain entry into the milieu of the neuronal tissue, their interaction with the ACE2 receptors in neurons could initiate a cycle of viral budding accompanied by neuronal damage. This can occur without substantial inflammation [
43]. In other cases, it may cause an acute inflammatory demyelination resulting in ADEM, which was described in one COVID-19 case [
30], and previously in MERS-CoV [
61]. COVID-19 is also thought to cause acute hemorrhagic encephalitis through the mechanism of a cytokine storm [
11]. Acute hemorrhagic leukoencephalitis is a rare demyelinating disorder that is usually fatal; ICU care, use of high-dose corticosteroid therapy, immunoglobulins, plasma exchange, and dehydrating agents have led to survival in only some patients [
62]. This has been seen in one COVID-19 case thus far [
11].
Three neuro-immunological entities related to COVID-19 infection surfaced in our systematic review: acute transverse myelitis, GBS and its variants, and Bell’s palsy [
9,
16,
28]. The time course of the disease is alike what is known for other viruses known to cause the above in that the symptoms of respiratory or gastrointestinal infection usually precede that of the neuro-immunological phenomena [
63]. This was seen in all cases in this report with the only exception being the GBS case report from China [
9]. This suggests that the underlying mechanism of such neuro-immunological phenomena in COVID-19 patients is likely to be grounded by the hypothesis of molecular mimicry, where mimicry between microbial and nerve antigens is thought to be a major driving force behind the development of the disorder [
63].
There are a number of limitations in our study. Most of the studies included in this systematic review are case reports/series and retrospective observational studies. The larger retrospective studies included were limited to the documentation of common neurological symptoms, such as headache, myalgia, and a loss in sense of taste and/or smell. Furthermore, the timeliness of our systematic review presents potential for premature analysis of data trends in the literature. Given the current pandemic, the authors of this study felt that the benefits of systematic data retrieval and review outweighed current risks of inaccurate predications. Our aim was to synthesize data to aid physicians currently treating the novel COVID-19, knowing time constraints inhibited their own analysis of the evolving literature on neurological manifestations.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.