Guillain-Barré Syndrome in a Patient With Evidence of Recent SARS-CoV-2 Infection

To the Editor:

A 58-year-old woman presented with rapidly progressive gait difficulty and dysgeusia after recovering from a febrile illness. Two weeks before presentation, she had returned from Florida but reported no contacts with persons who had confirmed or suspected coronavirus disease 2019 (COVID-19). She then developed an 11-day illness characterized by fever, myalgia, and asthenia but no respiratory symptoms (Figure ).

Figure.

Peripheral predominant opacities on chest CT (arrows) (A). T1 postgadolinium magnetic resonance images of the lumbar spine demonstrating smooth enhancement of the nerve roots (arrow) (B). Timeline showing progression of disease (C).

Six days after recovery, she noted dysgeusia without anosmia, followed by rapidly progressive bilateral paraparesis, imbalance, and severe lower thoracic pain without radiation. One week later, she was admitted locally because of progression of symptoms and now required a gait aid for ambulation. Results of a computed tomography angiogram of the chest and abdomen were negative for dissection but revealed peripheral predominant opacities (Figure). Laboratory work-up revealed a normal complete blood count and mild elevation in alanine aminotransferase at 73 U/L but otherwise normal liver function tests. She had an elevated D-dimer at 690 ng/mL, ferritin 575 μg/L, and sedimentation rate 26 mm/hour. Nasopharyngeal swab for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was negative by an emergency-use authorized real-time polymerase chain reaction (RT-PCR) test.¹

Given concern for COVID-19 despite the negative RT-PCR result, the patient was started on a 5-day course of hydroxychloroquine, zinc, and methylprednisolone 40 mg twice daily for 5 days, based on local hospital COVID-19 guidelines at that time. Because of progressive paraparesis and evolving areflexia, the local neurologist suspected Guillain-Barré syndrome (GBS). Cerebrospinal fluid (CSF) analysis revealed a protein of 273 mg/dL and 2 total nucleated cells; results of the meningitis/encephalitis panel were negative. Magnetic resonance imaging of the lumbar spine demonstrated smooth enhancement of the cauda equine roots (Figure). Results of locally performed anti-SARS-CoV-2 IgA and IgG serology (Euroimmun Inc., Lubeck, Germany) were positive. The patient was initiated on plasma exchange and received 1 treatment before transfer to our institution for further care.

Upon admission, cranial nerve examination—including olfaction—was normal. The patient had mild neck flexion weakness (Medical Research Council grade 4/5), mild/moderate (4/5) distal upper, and proximal and distal lower-limb weakness. Modified Erasmus GBS Outcome Score (mEGOS) was 1. Deep-tendon reflexes were absent in the legs and decreased in the upper extremities. Plantar responses were flexor. She had moderately severe length-dependent sensory loss in the feet, predominantly affecting large fiber modalities, and associated ataxic gait requiring 1-person assistance. Results of repeated nasopharyngeal SARS-CoV-2 RT-PCR were negative. Results of a qualitative SARS-CoV-2 IgG ELISA (Euroimmun) were again positive, with a signal to cutoff ratio (index value) of 8.2 (normal <0.8). Additional CSF studies included negative oligoclonal bands and IgG index as well as negative SARS-CoV-2 CSF RT-PCR and CSF/serum IgG antibody index results. Electrodiagnostic testing was performed, showing low amplitude and prolonged duration of the upper- and lower-limb compound muscle-action potentials, with prolongation of motor distal latencies, mild slowing of motor conduction velocities, and prolonged F-wave latencies but without conduction block or temporal dispersion on right-sided nerve conduction studies. Needle electromyography showed reduced recruitment of motor unit potentials in distal upper- and lower-limb muscles. These findings supported an acute sensorimotor demyelinating polyradiculoneuropathy, consistent with a diagnosis of GBS. Human immunodeficiency virus, syphilis, West Nile virus, and Lyme disease testing results were negative. Epstein Barr virus, Cytomegalovirus and Mycoplasma pneumoniae serology were consistent with remote infection. Ganglioside antibodies were negative as was serum and CSF paraneoplastic evaluation. She completed a total of 5 sessions of every-other-day plasma exchange. By dismissal, her motor and gait examination had improved. Although she remained slightly ataxic, she no longer required a gait aid.

Recently, cases of GBS have been reported in association with SARS-CoV-2 infection in China, Europe, and Iran.2, 3, 4, 5, 6, 7, 8, 9 In several, the onset of neurologic symptoms overlapped with active SARS-CoV-2 infection, suggesting a parainfectious process similar to that reported in association with Zika virus.^{10 , 11} Classic GBS is more commonly postinfectious, with symptoms developing 1 to 3 weeks after infection. This interval presumably permits the generation of antibodies that cross-react by molecular mimicry with specific components of peripheral nerves.¹² The cause of nerve injury in parainfectious cases is less clear, but direct damage from the virus or a hyperacute immune response have been postulated.

Our patient developed neurologic symptoms 17 days after the onset of fever (Figure). Results of real-time PCR for SARS-CoV-2 on a nasopharyngeal swab were negative, but the test was performed 3 weeks after onset of symptoms, at which point sensitivity is approximately 60% to 70%.¹³ Although the results of the CT chest scan were consistent with COVID-19 pneumonia, sputum or bronchoalveolar lavage SARS-CoV-2 RT-PCR was not pursued, given the absence of fever or cough. Overall, the temporal evolution of our patient suggests a postinfectious profile in the setting of probable SARS-CoV-2 infection. The results of the CSF SARS-CoV-2 RT-PCR and IgG antibody index were negative, arguing against neuroinvasion, but neither of these tests has been validated, and sensitivity is unknown. On electrodiagnostic testing, our patient had unequivocal demyelinating features, but both axonal and demyelinating variants have been described in association with SARS-CoV-2.^{3 , 4} Although scarce outcome data are available, our patient’s motor examination had substantially improved upon dismissal.

Cases of GBS are increasingly reported in the setting of SARS-CoV-2 infection. Given the ubiquity of the virus, coincident disease is a possibility, although the established association between GBS and infection argues against this proposition. Nonetheless, GBS appears to be relatively rare, based on the limited number of cases reported from countries past the peak of the first wave of the pandemic. It is possible, however, that the association was missed early on in the pandemic, either in critically ill patients who died of the illness or in patients with GBS who were not tested for the virus because of mild or no respiratory symptoms. Ongoing surveillance will be needed to confirm and further elucidate the nature of the association.