Introduction

Since the beginning of the COVID-19 outbreak, there has been a growing number of reports of Guillain–Barré syndrome (GBS) and Miller Fisher Syndrome (MFS) following severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, suggesting a post-infectious immune-mediated process [1]. However, the association between SARS-CoV-2 and GBS is still debated since a recent epidemiological study found no association between both entities [2]. In addition, pathophysiology remains unclear as there is no clear evidence of structural homology between SARS-CoV-2 and nerve compounds to support a molecular mimicry mechanism. Moreover, anti-ganglioside antibodies, which play a central role in pathogenesis—at least in the axonal forms—of GBS, have only been reported in a few cases [3]. Although there is no direct homology between SARS-CoV-2 structure proteins and any axonal or myelin surface proteins, it has been suggested that SARS-CoV-2 binds to respiratory tract gangliosides through its spike protein. Cross-reactivity between epitopes on SARS-CoV-2 spike-bound gangliosides and surface peripheral nerve glycolipids is currently considered as a pathophysiological hypothesis [4].

Methods

Between April and December 2020, four patients with a diagnosis of GBS according to the Brighton criteria, occurring after SARS-CoV-2 infection, were examined at Cliniques universitaires Saint-Luc, Universitair Ziekenhuis Brussel and Cliniques de l’Europe—Saint-Michel, three hospitals located in Brussels, Belgium. A positive diagnosis of COVID-19 infection was established by SARS-CoV-2 PCR assay of nasopharyngeal swab. Anti-ganglioside antibodies were tested by enzyme-linked immunosorbent assay (ELISA) for IgG and IgM antibodies against single gangliosides GM1, GM2, GM3, GM4, GD1a, GD1b, GD2, GD3, GT1a, GT1b, GQ1b and anti-sulfatides. Clinical and ancillary test descriptions were retrieved by the authors, who examined the patients. Consent was obtained from each patient for publication.

Results

Clinical characteristics and laboratory findings of the four patients with SARS-CoV-2-associated GBS are detailed in Table 1. The electrodiagnosis (EDX) findings are detailed in the Supplementary Appendix (S1-S4).

Table 1 Clinical characteristics and laboratory findings of four patients with SARS-CoV-2-associated Guillain–Barré syndrome
Full size table

All patients presented with sensorimotor symptoms and tendon areflexia about 3 weeks (20–24 days) after documented COVID-19 infection. SARS-CoV-2 infection was associated with a severe pulmonary disease in three of them. No other GBS triggering event was identified in all patients. CSF examinations revealed albuminocytologic dissociation in two patients (case1 and case2) and an increased albumin quotient in one (case4). Positive serum anti-GD1b antibodies were found in one patient (case4). The latter was the only patient showing brainstem involvement, with EDX (S4) compatible with an acute motor and sensory axonal neuropathy (AMSAN), and only partial clinical improvement after IVIG administration (2 g/kg for 5 days). This case was previously reported [5] but EDX was not available at that time. In the three other cases, the EDX (S1-S3) was consistent with acute inflammatory demyelinating polyneuropathy (AIDP) and no serum anti-ganglioside antibodies were detected. Two of the other patients (case1 and case2) presented severe SARS-CoV-2 infection and had good clinical outcome after IVIG treatment (2 g/kg for 5 days). The third other patient (case3) had mild SARS-CoV-2 infection and had spontaneously favorable evolution without treatment.

Discussion

We describe four cases of SARS-CoV-2-associated GBS, adding evidence to the probable association between these two entities. The lag time between COVID-19 infection and the neurological onset, the response to IVIG, and particularly the presence of anti-ganglioside antibodies in one patient are highly suggestive of a post-infectious immune-mediated mechanism.

Our patient with positive anti-ganglioside antibodies (anti-GD1b IgG) was the only one with brainstem involvement, suggestive of GBS/Bickerstaff brainstem encephalitis (BBE) overlap, and with electrophysiological features suggestive of AMSAN, a previously reported association [6]. The three other patients presented with AIDP and had negative anti-ganglioside screening. While our patient presented myoclonus limited to the palatal region, myoclonus has already been described in BBE [7]. Anti-GD1b have been described in the clinical spectrum of GBS—classically in acute sensory ataxic neuropathy (ASAN), but also in AMSAN—, MFS and BBE, and are associated with more severe disease and slower recovery [5, 8, 9]. Although, the 7-week delay of EDX raises the question of a probable critical illness polyneuropathy component—which is difficult to distinguish from an AMSAN—, the other features and the response to IVIG are in favor of a post-infectious immune-mediated process.

According to a recent review of the literature, anti-ganglioside antibodies in COVID-19-associated GBS were positive in only 5/36 patients (14%) [3]. Four new anti-ganglioside positive cases, including one of our patients, were identified reviewing the literature for this paper and are summarized in Table 2. In COVID-19 patients, anti-ganglioside antibodies were found not only in patients with GBS but also in patients with variable neurological presentations (cranial neuropathy with meningo-polyradiculitis, choreic movements, myelitis) [5, 8,9,10] which casts doubt on whether these are truly pathogenic in all cases. Interestingly, anti-ganglioside antibodies have been described in other neurological diseases besides the GBS/MFS/BBE spectrum, even though their clinical significance remains unsure [11]. On the other hand, the role of anti-ganglioside antibodies in the genesis of axonal subtypes of GBS after Campylobacter jejuni infection is well established, and associated with severe disability [12].

Table 2 Anti-ganglioside antibodies in SARS-CoV-2-associated neurological syndromes
Full size table

Prevalence of anti-ganglioside antibodies in GBS was, respectively, 31.7% and 50% in two large cohorts of, respectively, 306 and 119 patients [13, 14]. The current low rate of positive anti-ganglioside antibodies in SARS-CoV-2-associated GBS could be underestimated due to lack of testing, or due to the fact that a majority of reported SARS-CoV-2-associated GBS are of the AIDP variant (48/62 [77.4%];40/75 [75%]) [1, 15], which is known to be less frequently associated with serum anti-ganglioside antibodies than the axonal forms [12, 13, 15,16,17]. Another explanation might be the presence of antibodies against untested gangliosides, ganglioside complexes or other more atypical antigens. Anti-ganglioside complex antibodies are difficult to detect and are not tested in routine clinical practice but can be useful in patient with negative antibodies against single gangliosides [18]. It may be hypothesized that the SARS-CoV-2 spike-bearing gangliosides could form a complexed neoantigen recognized by unidentified specific anti-complex antibodies. On the other hand, antibodies less typically associated with GBS could also have a pathophysiological role. Positive anti-CASPR2 antibodies have been reported in rare cases of GBS [9], and one patient suffering from an encephalitis following SARS-CoV-2 infection had positive anti-CASPR2 antibodies [5]. Currently, anti-CASPR2 antibodies are not routinely screened in GBS patients and might thus be underreported.

Finally, other mechanisms such as the ‘cytokine storm’ could hypothetically be involved in the pathogenesis of SARS-CoV-2-associated GBS, especially in the acute onset within days after onset of viral infection, since many of the released cytokines have been implicated in the pathogenesis of ‘classical’ GBS [19].

Conclusion

We report four cases of SARS-CoV-2-associated GBS. The interval of 3 weeks between SARS-CoV-2 symptoms and neurological onset, the favorable response to IVIG, and the positive anti-ganglioside in one patient further support the hypothesis of an immune-mediated post-infectious process. Further autoantibody testing (including anti-ganglioside, anti-neuronal antibodies like anti-CASPR2, or anti-complex antibodies) in SARS-CoV-2 patients exhibiting neurological symptoms could allow a better understanding of these probable immune-mediated post-infectious processes.