Miller Fisher Syndrome in Patients With Severe Acute Respiratory Syndrome Coronavirus 2 Infection: A Systematic Review

Miller Fisher syndrome (MFS) is recognized as a rare variant of Guillain-Barré syndrome (GBS) and defined by the acute onset of the triad of ophthalmoparesis, areflexia, and ataxia.1 There is evidence of MFS being preceded by infections similar to those preceding GBS.2 Cases of MFS3, 4 have recently been linked to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, which is the causal agent of coronavirus disease 2019 (COVID-19), but no studies have synthesized the characteristics of patients with this condition, which is critical for clinical practice. Here we systematically describe the clinical features of MFS in patients with COVID-19.

In this rapid systematic review, we searched the PubMed database to identify studies of cases of MFS associated with SARS-CoV-2 infection. We included studies provided clinical data and information on neurological examinations, cerebrospinal fluid (CSF) analyses, and antiganglioside antibody tests. We excluded cases in which SARS-CoV-2 infection was not confirmed using a polymerase chain reaction test.

Reports were screened in two stages: 1) screening of titles and Abstracts, followed by 2) the retrieval and screening of full-text articles. PubMed was searched from January 1, 2020 up to January 30, 2021, without language restrictions. The reference lists for all eligible studies and reviews were also evaluated to identify additional studies for inclusion. We used the following search strategy for primary studies: (“Miller Fisher Syndrome” [MeSH] OR “Fisher Syndrome” OR “Miller-Fisher”) AND (LitCGeneral [Filter]). Data were extracted from publications by two authors and cross-checked for accuracy. The results were collated in a descriptive manner.

After screening 44 titles and Abstracts, 12 full-text articles3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 were assessed for eligibility, resulting in the exclusion of 3 studies.6, 7, 14 Two additional studies15, 16 were identified from the reference lists, and finally 11 studies3, 4, 5, 8, 9, 10, 11, 12, 13, 15, 16 with casereport designs were included in this review (Fig. 1). The patients were aged from 31 to 74 years (median 51 years) and most of them were male (n=8, 73%). In three (27%) cases,5, 10, 11 patients were hospitalized with acute motor or sensory polyneuropathy as the first sign of SARS-CoV-2 infection. Two of these patients presented paresthesia without motor weakness, and the third reported upper-limb weakness without sensory symptoms. In the remaining cases (75%),3, 4, 8, 9, 12, 13, 15, 16 neurological manifestations were reported up to 3 weeks after typical COVID-19 symptoms (Table 1).

Fig. 1
Flow diagram of studies included in the systematic review.

• Click for larger image
• Download as PowerPoint slide

Table 1
COVID-19 symptoms and the onset of neurological manifestations in patients with Miller Fisher syndrome

• Click for larger image
• Click for full table
• Download as Excel file

The most common neurological feature was hyporeflexia or areflexia (100%), followed by ataxia (91%), ophthalmoparesis/diplopia (82%), sensory symptoms (73%), weakness of facial muscles (55%), and eyelid ptosis (36%). Magnetic resonance imaging was performed for nine patients,4, 5, 8, 10, 11, 12, 13, 15, 16 but only one (11%) case8 presented alterations, which were characterized by enhanced T2-weighted hyperintensity and enlargement of the oculomotor nerve. Neurophysiological studies were performed in five patients,4, 9, 12, 15, 16 with acute inflammatory demyelinating polyneuropathy diagnosed in four (80%) of them.9, 12, 15, 16 CSF investigations were described for nine patients,3, 4, 5, 9, 10, 11, 12, 13, 16 which revealed albuminocytological dissociation in most cases (78%).3, 4, 5, 9, 11, 12, 13 The CSF analyses did not produce any positive results for SARS-CoV-2 RNA. A ganglioside antibody panel was explored for nine patients,3, 4, 8, 9, 10, 11, 12, 13, 16 with the presence of anti-GD1b3 and anti-GQ1b10 found in only two (22%) cases. Ten (91%) patients were treated with intravenous immunoglobulin (IVIg), which resulted in clinical improvement in nine patients; the tenth patient11 died of ventricular arrhythmia as manifestation of dysautonomia 2 weeks after the onset of neurological manifestations. One patient did not receive any pharmacological treatment (Table 2).

Table 2
Clinical, laboratory, and neuroimaging findings for patients with Miller Fisher syndrome

• Click for larger image
• Click for full table
• Download as Excel file

Neurological manifestations are common in COVID-19 and they may represent the only disease symptom.17 There is a potential association between SARS-CoV-2 infection and neurological symptoms, but the underlying biological mechanisms remain poorly defined.18 It was recently found that angiotensin-converting enzyme 2 may be expressed in neurons, astrocytes, oligodendrocytes, and the olfactory bulb, which is critical for SARS-CoV-2 cellular tropism in humans.19 There is increasing evidence of the neuroinvasivity of SARS-CoV-2 in postmortem examinations20, 21 and the viral RNA detection in CSF samples from patients with meningitis22 and encephalitis.23 However, a postinfectious immune-mediated process has been speculated as the main mechanism of SARS-CoV-2-associated neuropathology.24

Most of the patients with MFS in this study presented albuminocytological dissociation (CSF with total white cell count <10 cells/µL and protein above the normal laboratory range). SARS-CoV-2 RNA was not detected in the CSF, suggesting a mechanism of immune-mediated injury rather than direct viral neurotropism. There is emerging evidence that immune-mediated inflammatory mechanisms is associated with the development of neurological disorders in patients with COVID-19 due to the increased release of cytokines and chemokines, including IL-1, IL-6, IL-17, IL-22, and TNF-α.25, 26, 27 These observations have important clinical implications for the treatment of MFS associated with COVID-19. In clinical practice, patients with MFS exhibit good clinical outcomes when treated with IVIg and plasma exchange. This systematic review found no difference in the prognosis between anti-GQ1b-positive and -negative patients, with almost all of them presenting at least partial improvement during the first 2 weeks of treatment, which supports the immune-mediated injury mechanism. One patient who was negative for anti-GQ1b showed initial improvement on the second day of IVIg but died a few days later due to ventricular arrhythmia. Only one patient did not receive any pharmacological treatment, which was due to their MFS symptoms being considered mild.

Moreover, the presence of antibodies against gangliosides—a classical feature found in patients with non-COVID-19 MFS28, 29, 30—does not seem to be a useful diagnostic marker for MFS associated with COVID-19. Anti-GQ1b antibody has been reported to be present in 81% of patients with MFS.31 Despite gangliosides being a possible target for IgG antibodies for patients exposed to viral infections,2, 30, 32 the presence of antiganglioside antibodies in patients with MFS associated with SARS-CoV-2 infection seems to be uncommon. In MFS patients who are positive for anti-GQ1b antibodies, there is a possibility of cross-reaction with other gangliosides such as GT1a, GD1b, and GD3. GD3 and arginylglycylaspartic acid (RGD) are known to be involved in cell adhesion,33 and RGD has been suggested as an alternative receptor for SARS-CoV-2.34 We therefore hypothesized that different targets and immune-mediated mechanisms could be associated with the neuropathology of these patients.

In conclusion, this study has synthesized the published literature on MFS in patients with SARS-CoV-2 infection. Although rare, patients with COVID-19 may present neurological symptoms suggestive of MFS. SARS-CoV-2 RNA was not detected in the CSF analyses. The presence of antibodies against gangliosides was uncommon, and almost all patients exhibited good clinical outcomes after treatment with IVIg.