Post-Infectious Guillain–Barré Syndrome Related to SARS-CoV-2 Infection: A Systematic Review

Background. Guillain-Barré syndrome (GBS) is the most common cause of flaccid paralysis, with about 100,000 people developing the disorder every year worldwide. Recently, the incidence of GBS has increased during the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) epidemics. We reviewed the literature to give a comprehensive overview of the demographic characteristics, clinical features, diagnostic investigations, and outcome of SARS-CoV-2-related GBS patients. Methods. Embase, MEDLINE, Google Scholar, and Cochrane Central Trials Register were systematically searched on 24 September 2020 for studies reporting on GBS secondary to COVID-19. Results. We identified 63 articles; we included 32 studies in our review. A total of 41 GBS cases with a confirmed or probable COVID-19 infection were reported: 26 of them were single case reports and 6 case series. Published studies on SARS-CoV-2-related GBS typically report a classic sensorimotor type of GBS often with a demyelinating electrophysiological subtype. Miller Fisher syndrome was reported in a quarter of the cases. In 78.1% of the cases, the response to immunomodulating therapy is favourable. The disease course is frequently severe and about one-third of the patients with SARS-CoV-2-associated GBS requires mechanical ventilation and Intensive Care Unit (ICU) admission. Rarely the outcome is poor or even fatal (10.8% of the cases). Conclusion. Clinical presentation, course, response to treatment, and outcome are similar in SARS-CoV-2-associated GBS and GBS due to other triggers.


Introduction
Guillain-Barré syndrome (GBS) is the most common cause of flaccid paralysis, with about 100,000 people developing the disorder every year worldwide [1]. GBS is considered a postinfectious disease as approximately two-thirds of patients report preceding infections with specific pathogens, such as Campylobacter jejuni (32%), cytomegalovirus (13%), and Epstein-Barr virus (10%) [2].
Recently, several cases of GBS were reported during the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) epidemics worldwide. Patients with COVID-19 typically have fever and respiratory illness; however, a wide range of other symptoms have been described. While the neurological sequelae of the virus remain poorly understood, there are a growing number of reports of neurological manifestations of COVID-19 [3].
GBS is an acute, immune-mediated polyradiculoneuropathy with a wide range of clinical manifestations. The classic form of GBS is characterized by a rapidly progressive

Protocol and Registration
We performed a systematic review based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [7]. The protocol was not published, and the review was not registered with the International prospective register of systematic reviews (PROSPERO).

Literature Search
We identified the articles by searching electronic databases (Embase, MEDLINE, Google Scholar, and Cochrane Central Trials Register). Other relevant studies were identified from the reference lists. We used a combination of such terms as "Guillain-Barré Syndrome" and "COVID-19". The initial search was performed on 24 September 2020. The titles and abstracts were screened by two researchers (PS and LGG) to identify the key words. The selected papers were read in full by the two independent reviewers and a third reviewer (MCP) was consulted in case of disagreement.
We included all the papers with available full text, without any restriction of the year of publication, reporting original data of patients with GBS and a suspected, probable, or confirmed recent SARS-CoV-2 infection, of any age, gender, and in any setting.
Predefined exclusion criteria were: GBS within 3 months after a vaccination or other proven triggering infection (e.g., C. jejuni), and studies with no information on residence, and at least one clinical variable of interest. Cases were classified according to the reported diagnostic certainty levels for GBS and SARS-CoV-2 infection. To classify the diagnosis GBS, we employed the "Brighton Collaboration Criteria (2011)" [8]. These were defined on the basis of the available reported data. The diagnostic certainty of SARS-CoV-2 infection was established according to the World Health Organization (WHO) criteria [9].

Data Extraction and Management
Clinical characteristics were reported together with the ratio of the number of patients in whom the variable was present (n) and the total number of reported cases (N): n/N (%). As for the symptoms, we assumed they were absent rather than missing if they were not cited in the manuscript, in order to account for the reporting bias, and therefore described as zero (n) out of the total number of reported cases (N).
Continuous variables (age, time between infectious and neurological symptoms, duration of progression and plateau phase of GBS, duration of hospital admission) were extracted as medians and or means, depending on how they were presented in the original article.

Study Selection
We identified 63 articles in the researched databases, and 32 of them were included in our systematic review. The 32 selected studies reported on a total of 41 GBS cases with a confirmed or probable COVID-19 infection: 26 of them were single case reports and 6 case series. The flow diagram (see Figure 1) shows the results from the literature search and the study selection process.

Study Characteristics
All the Studies Are Presented Alphabetically with a Brief Clinical Description Per Case (Table 1).  Recovery.

F, 71
Cough and fever.
Dysarthria, progressive weakness and numbness of the lower extremities.
Asthenia, hands and feet paresthesia and gait difficulties.      China F, 61 Cough and fever.
Progressive limb weakness and foot dysaesthesia.

Mild improve ment.
Zhao H et al., 2020 [41] China F, 61 Cough and fever. All the cases were from COVID-19 epidemic or endemic regions (see Figure 2): 12 cases were from Italy (29.3%), 7 from U.S. (17.1%), 6 from Spain (14.6%), 4 from France (9.7%), 3 from Iran (7.3%) and U.K. (7.3%), 2 from Germany (4.9%), and one from China (2.4%), Morocco (2.4%), Switzerland (2.4%), and Turkey (2.4%). Thirty-three cases were positive for nasopharyngeal swab test for SARS-CoV-2 by qualitative RT-PCR assay, four for oropharyngeal swab test for SARS-CoV-2 by qualitative RT-PCR assay, two for anti-SARS-CoV-2 IgA, and four for anti-SARS-CoV-2 IgG. In Manganotti P et al., the patient was reported to be COVID-19 positive with no further information provided. After discussion, we decided to consider it as a probable case of SARS-CoV-2 infection. Thirty-three cases were positive for nasopharyngeal swab test for SARS-CoV-2 by qualitative RT-PCR assay, four for oropharyngeal swab test for SARS-CoV-2 by qualitative RT-PCR assay, two for anti-SARS-CoV-2 IgA, and four for anti-SARS-CoV-2 IgG. In Manganotti P et al., the patient was reported to be COVID-19 positive with no further information provided. After discussion, we decided to consider it as a probable case of SARS-CoV-2 infection.
A chest CT scan was performed in fifteen cases showing multiple bilateral ground glass opacities, typical of COVID-19 pneumonia. Eleven studies reported a chest X-ray: five studies showed bilateral ground glass pneumonia. The classic form of GBS was diagnosed in 33 cases. Twenty-one of the twenty-seven cases where the Brighton classification was reported, fulfilled level 1. The most frequent clinical phenotype was demyelinating polyradiculoneuropathy. In 8 cases, the variant MFS was diagnosed. SARS-CoV-2 infection was confirmed in 40 (97.6%) and probable in 1 (2.4%) of all cases.

Clinical Characteristics
All studies reported the presence of clinical symptoms of SARS-CoV-2 infection (see Figure 3). Two or more symptoms were present in 87.5% of cases. The most common symptoms were fever (63.4%, 26/41 cases), cough ( A chest CT scan was performed in fifteen cases showing multiple bilateral ground glass opacities, typical of COVID-19 pneumonia. Eleven studies reported a chest X-ray: five studies showed bilateral ground glass pneumonia.
The classic form of GBS was diagnosed in 33 cases. Twenty-one of the twenty-seven cases where the Brighton classification was reported, fulfilled level 1. The most frequent clinical phenotype was demyelinating polyradiculoneuropathy. In 8 cases, the variant MFS was diagnosed.

Demographics
The median age of the study populations varied between 36 and 74 years, and only 1 pediatric patient was included in Paybast S et al. The majority of patients were male (62.8%) and the male:female ratio of all studies combined was 1.69. Toscano G et al. did not report the patients' age. SARS-CoV-2 infection was confirmed in 40 (97.6%) and probable in 1 (2.4%) of all cases.

Diagnostic Investigations
PCR, principally from nasal or throat swab, was the most frequently performed test for SARS-CoV-2 diagnosis. Anti-SARS-CoV-2 IgA was positive in two cases (Coen M et al., and Naddaf E et al.), and anti-SARS-CoV-2 IgG in three cases (Coen M et al., Naddaf E et al., and Riva N et al.). In the CSF, SARS-CoV-2 PCR was negative in all tested cases.
Only eleven studies tested all the cases for other infections that have been associated with GBS (C. jejuni, CMV, EBV, hepatitis E virus, mycoplasma pneumoniae) and all the tested cases (11/41; 26.8%) were negative for recent infection. None of the studies tested for all of these pathogens.
CSF was examined in thirty-one studies, and information on protein level and cell count was provided by all of these. Increased protein level and albuminocytological dissociation were present in the vast majority of cases. Only three studies reported normal CSF cell count: 1 case in Oguz-Akarsu E et al., 1 case in Riva N et al., and two cases in Toscano G et al.
Electrophysiological studies were done in 30 (73.2%) of the reported cases. The most frequent electrophysiological subtype was AIDP in 56.7%, followed by AMSAN in 16.7% and AMAN in 13.3% of cases. In three studies (Ottaviani D et al., Paybast S et al., and Rana S et al.) electrophysiological studies showed a mixed pattern of demyelination and axonal damage. The investigations are reported in Table 3.

Treatment and Disease Progression
As shown in Table 3

Discussion
It is now known that SARS-CoV-2 can often affect central and peripheral nervous system, apart from the bronchopulmonary system [42].
The current study reports various cases of GBS or its variants in patients with SARS-CoV2 infection. Our systematic review shows that published studies on SARS-CoV-2related GBS typically report a classic sensorimotor type of GBS often with a facial palsy and a demyelinating electrophysiological subtype. The time between onset of infectious and neurological symptoms and negative PCR in most patients suggests a post-infectious mechanism rather than a direct infectious one.
The mechanism involving the peripheral nervous system in SARS-CoV-2-infected patients and the development of GBS is unknown. Since most patients developed GBS on average 10 days after the first non-neurological symptoms of the SARS-CoV-2 infection, a causal relation is quite likely.
However, in none of the included patients, specific SARS-CoV2 RNA was found in the CSF. This suggests that GBS is not triggered by a direct viral attack against the nerve roots but rather by an immune-mediated mechanism, such as antibody precipitation on myelin sheaths or axons [43]. COVID-19 is associated with a cytokine storm and a dysregulated immune response [44]. A mimicry between the epitopes on the surface of the virus and the ones on neuronal membranes is possible, because they are simultaneously attacked by the immune response, just like in the GBS due to C. jejunii [45]. It is also possible that the virus directly invades motor and sensory neurons since the presence of the virus was reported in neural and capillary endothelial cells in frontal lobe tissue obtained at post-mortem examination from a patient infected with SARS-CoV-2 [46].
Published studies on SARS-CoV-2-related GBS typically report a classic sensorimotor type of GBS often with a demyelinating electrophysiological subtype. The distribution of subtypes varies between countries. In Europe and the United States, AIDP affects 60-80% of people with GBS, while AMAN affects only a 6-7%. In Asia and Central and South America, that proportion is nearly 30-65%. This is related to the exposure to various infections and to the genetic characteristics of population [47]. This study found that in Western countries, the demyelinating subtype affected most of the people with GBS, while AMAN/AMSAN affected only a small number. MFS was reported in a quarter of the cases.
A male preponderance was noted as with non-SARS-CoV-2-associated GBS. Similarly, elderly patients are more frequently affected than the younger.
We noticed how GBS can arise in patients who are totally asymptomatic or with moderate symptoms of COVID-19. Furthermore, the severity of GBS is not correlated with that of COVID-19. GBS-related neurological symptoms have often overlapped with COVID19-related symptoms, due to the short time between the SARS-CoV-2 infection and the onset of GBS (8-23 days). It is therefore evident that this overlap contributes to a poor prognosis.
Increased protein level and albuminocytological dissociation were present in most cases. This pattern distinguishes GBS from other conditions, such as lymphoma and poliomyelitis, in which both the protein and the cell count are elevated. Furthermore, protein level in CSF was shown to correlate with disease activity, progression, and response to the treatment [48].
None of the patients with reported CSF analysis had SARS-CoV-2 in the CSF. This is in contradiction with various case reports, which detected SARS-CoV-2 in the CSF [49,50]. A possible explanation is the lack of disruption of the blood-brain barrier that would allow SARS-CoV-2 to cross the CSF space [51].
Another possible explanation is the low sensitivity, around 60%, of the currently available real-time reverse-transcription polymerase chain reaction (rRT-PCR) [52].
In almost all cases, the response to immunomodulating therapy was favorable. Intravenous immunoglobulins and plasmapheresis are the two main immunotherapy treatments for GBS [53,54]. A five-day course of daily IVIg (0.4 g/kg/day) was the most common regime adopted. Plasma exchange was performed four times. In two cases, IVIg and PLEX were both administered but a combination of the two is not significantly better than either alone [53]. Only a patient received corticosteroids, even if it is shown that they are not effective in speeding recovery and could potentially delay recovery [55].
The disease course was frequently severe and about one-third of the patients with SARS-CoV-2-associated GBS required mechanical ventilation and ICU admission. Rarely the outcome was poor or even fatal.
Limitations. Our study has several limitations. First, the studies included in this systematic review are variable in study design and setting, selection criteria, diagnostic ascertainment, and reporting of variables, which are potential sources of bias. Second, given the concurrent manifestations of COVID-19 and GBS, it is possible that some COVID-19 symptoms could be attributed to GBS and vice versa, since both diseases affect the respiratory system. Third, such cases without results of nerve conduction were not categorized into GBS subtypes, since it was not possible to confirm whether the symptoms were due to demyelination or axonal damage.

Conclusions
The SARS-CoV-2 infection can cause GBS. The underlying mechanism leading to development of this condition is still unclear. The CSF is free of SARS-CoV2 RNA, therefore excluding a viral attack directed against the nerve roots, but various hypotheses are possible. Clinical presentation, course, response to treatment, and outcome are similar in SARS-CoV-2-associated GBS and GBS due to other triggers. It is important to be aware of this association to avoid delay in diagnosis and to promote early treatment initiation given the significant risk of morbidity and mortality.
Author Contributions: P.S. and L.G.G. helped design the study, conduct the study, analyse the data, and writethe manuscript; C.A., F.C., V.E., M.F., A.P., M.B.P. and V.P. helped design thestudy and analyse the data; M.C.P. helped designthe study, conduct the study and analyse the data. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.