Acute Cerebellitis and Myeloradiculitis Associated With SARS-CoV-2 Infection in Common Variable Immunodeficiency—A Case Report

Abstract

The role of the adaptive immune system in mediating COVID-19 is largely unknown. Therefore, it is difficult to predict the clinical course in patients with common variable immunodeficiency (CVID), a disease characterized by dysfunctional lymphocytes and impaired antibody production. We report a case of SARS-CoV-2 infection presenting as isolated neurological symptoms in a patient with CVID. The patient subsequently improved following steroids, intravenous immunoglobulin, and convalescent plasma (CP). The latter has been shown to be safe and efficacious in treating COVID-19 in patients with primary immunodeficiency. Recent data suggest that the mechanism of CNS injury in COVID-19 may be due to immunological dysregulation rather than direct viral-mediated injury. This case exemplifies the complex interaction between the brain, the immune system, and the SARS-CoV-2 virus.

Introduction

Since its emergence in 2019, the SARS-CoV-2 virus has been associated with neurological disease, the full breadth of which has yet to be discovered. ¹ We present the case of a patient with common variable immunodeficiency (CVID), the most prevalent form of severe antibody deficiency characterized by impaired B-cell differentiation with low serum levels of IgG, IgA, and/or IgM. ² She presented with an acute cerebellar syndrome and myeloradiculopathy in the setting of mild COVID-19 disease. She experienced subsequent resolution of her neurological symptoms after administration of intravenous immunoglobulin (IVIg), steroids, and convalescent plasma (CP).

Case Presentation

A 54-year-old woman with a history of common variable immunodeficiency (CVID), complicated by recurrent afebrile urinary tract infections (UTIs), pneumonia, and breast cancer in remission (on tamoxifen), presented with difficulty walking and bilateral upper-extremity tremors worsening over the course of 1 week. The patient was diagnosed with CVID at age 28 after recurrent upper and lower respiratory infections. At that time, she was noted to have low levels of serum Ig: IgG = 251 mg/dl (normal range = 768–1728 mg/dl), IgA = 7 mg/dl (normal range = 99–396 mg/dl), and IgM = 20 mg/dl (normal range = 38–266 mg/dl). She had no respiratory or generalized systemic symptoms and denied recent travel or sick contacts. The patient had received monthly maintenance infusions of 40 g IVIg for CVID, with her last infusion given on schedule 3 weeks prior to presentation. Her home medications also included metoprolol and hydrochlorothiazide.

On examination, the patient was well-appearing and with normal vital signs. Her neurologic exam was notable for marked dysmetria and intermittent myoclonus in all of her extremities, prominent truncal unsteadiness, hyperreflexia in the upper extremities, and a wide-based ataxic gait. The rest of her neurologic exam was unremarkable. An MRI of the brain revealed symmetric, non-enhancing T2 white matter hyperintensities in the bilateral striatum and inferior cerebellar hemispheres (Figures 1A and B). A routine SARS-CoV-2 nasopharyngeal swab was positive. Chest CT was negative for SARS-CoV-2 pneumonia (Figure 1E).

Figure 1.

Imaging of the index patient. (A) Axial diffusion weighted imaging and (B) Fluid-attenuated inversion recovery MRI on admission showing hyperintensities in striatum and cerebellar hemispheres (red arrows). (C) Axial diffusion weighted imaging and (D) fluid-attenuated inversion recovery MRI performed 3 months after discharge showing complete resolution of hyperintensities. (E) CT of the chest showing normal pulmonary parenchyma. (F) Axial and (G) sagittal MRI of the spine with gadolinium showing enhancement of the cauda equina (red arrows) at the level of the 12th thoracic vertebrae (red star).

An MRI of the spine demonstrated abnormal enhancement of the cauda equina (Figures 1F and G). Complete blood count (white blood cells = 6.2 K/μL, 78% neutrophils, 11% lymphocytes, 7% monocytes, and 0.5 eosinophils), comprehensive metabolic panel, and thyroid function tests were normal. A lumbar puncture revealed CSF WBC of 12/μL and protein of 44 mg/dL. The remaining CSF studies were negative, including glucose, cultures, oligoclonal bands, cytology, and flow cytometry. SARS-CoV-2 CSF PCR was negative. Extensive testing for infectious, autoimmune, paraneoplastic, metabolic, and toxic etiologies were unrevealing (Table 1). CT of the chest, abdomen, and pelvis showed no evidence of malignancy.

Table 1.

Work-Up of the Index Patient.

Category	Tests Performed
Infectious	Bacterial cultures, herpes simplex virus, varicella-zoster virus, cytomegalovirus, Epstein–Barr virus, West Nile virus, JC virus, lyme, and cryptococcus
Autoimmune	Anti–double-stranded DNA, anti-nuclear antibody, angiotensin-converting enzyme
Serum/CSF paraneoplastic	AMPA-R, CASPR2, GABA-B-R, GAD65, LGI-1, NMDA-R, anti-Hu, anti-Yo, and anti-Ri
Metabolic	B12, copper, PTH, vitamin D, and carboxyhemoglobin

High-dose intravenous steroids (methylprednisolone, 1 g IV daily for 5 days) were administered, and only mild clinical improvement was seen on day 5 of admission. The patient reported mild improvement of the myoclonus; however, there was significant gait instability and no changes in the ataxia. CP was then administered to augment viral clearance. Following two infusions of CP (200 mL each; day 5 and 13), the patient experienced a marked improvement in gait and appendicular ataxia, and resolution of myoclonus on day 9 of admission (Figure 2). She also received her usual maintenance IVIg infusion during her hospital course on day 5 following treatment with CP. Notably, the levels of SARS-CoV-2 IgG became positive on the fifth day of admission (before administration of IVIg). She continued to improve rapidly and was discharged home requiring only a cane to mobilize; however, she continued to test positive on serial nasopharyngeal SARS-CoV-2 PCR throughout the hospitalization and at the time of discharge (day 16). At 3 months follow-up, she had minimal residual spasticity and her walking returned back to baseline. There was a radiologic resolution of the T2 hyperintensities previously seen in the basal ganglia and cerebellum (Figures 1C and D).

Figure 2.

Hospital course of the index patient. Hospital course and follow-up of the index patient. Black arrows correspond to the dates that an MRI was performed. Orange arrows indicate the levels of SARS-CoV-2 IgG; negative values are represented by empty arrows, and positive values are represented by filled arrows (negative are levels < 5 absorbance units/mL); * = positive levels of SARS-CoV-2-IgM (not shown) were obtained on day 11. Green arrows represent the days SARS-CoV-2 nasopharyngeal swab was performed; +, positive; and NP, nasopharyngeal. The gray square represents a five-day course of high-dose steroid (methylprednisolone 1 g IV daily for 5 days). Blue arrows correspond to the dates of convalescent plasma administration (200 mL per dose). The purple arrow corresponds to the day that intravenous immunoglobulin infusion was administered. Red rectangles correspond to the clinical improvement seen; the thickness of the rectangle represents the degree of clinical improvement; IV, intravenous; Ig, immunoglobulin; D, days; and m, months.

Discussion

Herein, we report a case of isolated neurological manifestations associated with SARS-CoV-2 infection with unique radiological findings in a patient with CVID, with subsequent improvement following CP. More than 35% of patients with COVID-19 report neurological symptoms. ¹ Rare cases of cerebellitis, ³ myelitis, or polyradiculoneuropathy in the setting of COVID-19 have been reported.^4,5 Neurological symptoms appear to have a higher prevalence in severe illness ⁶ ; however, it is notable that purely neurological manifestations with mild or absent systemic symptoms have been reported in patients without impaired immunity. ⁵ While patients with CVID have been seen to have a relatively mild course with SARS-CoV-2 infection, ⁷ to our knowledge, this is the first report of a CVID patient developing primarily neurologic disease in the setting of otherwise asymptomatic COVID-19 disease.

Recent studies have suggested that pathogen–host interactions lead to a dysregulated host inflammatory response that may contribute to the virulence and systemic sequelae of patients infected with SARS-CoV-2. ⁸ In the central nervous system (CNS), the removal of the virus from the tissues by the innate and adaptative immune responses could result in neurological damage. ⁹ The innate immune system has been shown to have a major role in the acute phase of infection. Namely, immune effector cells have been shown to release pro-inflammatory cytokines and chemokines, a cocktail that potentiates further systemic inflammation and may ultimately culminate into a cytokine storm.^10,11

The role of the adaptive immune system in mediating SARS-CoV-2 infection is less studied. ¹² It is therefore difficult to predict how the disease would behave in patients with CVID who have dysfunctional lymphocytes and lack of antibody production. Interestingly, iatrogenic B-cell deficiency secondary to rituximab appears to be associated with more severe COVID-19. ¹³ A study of 135 patients with CVID found that of ten patients who were infected with SARS-CoV-2, only one patient required hospitalization, whilst the rest tended toward a milder and shorter clinical course. ⁷ Another study of patients with primary immunodeficiencies infected with COVID-19 reported that seven of nine patients recovered, whilst only two patients died, both of whom had pre-existing pulmonary disease. ¹⁴ In contrast, in the general population, trends in COVID-19 illness are notable for: 14% hospitalized, 2% admitted to ICU, and 5% dying. ¹⁵ These observations suggest the role of B cells in the course of the infection is not so straightforward. ⁷ One study of serum and CSF antibody titers in patients with neurological symptoms found that about half of patients were positive for SARS-CoV-2–specific antibodies in CSF; the antibody titer was associated with disease severity. ¹⁶ Although further data are required to draw conclusions, there may indeed be a role for B cells in perpetuating the rampant inflammatory response, which may underlie the absence of typical respiratory and systemic symptoms in this immunocompromised patient.

To add a further layer of complexity, the pathological effects of SARS-CoV-2 in the CNS are yet to be fully described. Direct inflammation to the CNS has been questioned by two neuropathologic studies showing multifocal microvascular injury and hypoxic changes only in the brains of those infected with SARS-CoV-2.^17,18 Microvascular ischemic injury has also been hypothesized to be the etiology of patients with myelopathy that lack a response to immunomodulatory agents and have a poor recovery. ⁵ On the contrary, one post-mortem study of brain tissue from 110 COVID-19 patients found evidence of a CNS inflammatory response with glial activation and perivascular cytotoxic T cell infiltration. ¹⁹ They observed microglial activation with occasional microglial nodules, a finding seen in the setting of autoimmune encephalitis, ²⁰ which was most pronounced in the brainstem and cerebellum. ¹⁹ Additionally, SARS-CoV-2 could be detected by PCR in the brain tissue of most patients; however, the presence of SARS-CoV-2 did not correlate with the severity of the histological changes observed. ¹⁹ This comprehensive study sheds light on cellular mechanisms underpinning CNS injury in SARS-CoV-2 infection, suggesting a possible immunological basis for the observed neuropathology, and less likely direct viral-mediated injury.

Our patient demonstrated an early and dramatic response to steroids, IV Ig, and CP. CP is a therapy that is controversial in SARS-CoV-2 infection. The therapeutic mechanism of CP is not fully understood. Recent literature suggests that CP reduces the severity of the disease through a multifaceted approach: directly neutralizing the virus, enhancing humoral immune response, and inhibition of natural killer cells and complement. ²¹ Disappointingly, a recent randomized clinical trial demonstrated no difference in clinical improvement or mortality amongst patients with severe pneumonia treated with CP vs placebo. ²² In patients with immunodeficiency, however, CP has been reported to be a safe and efficacious therapy for COVID-19.^23,24 Our patient had a similar outcome, although she initially presented with an incredibly unusual presentation. Overall, we acknowledge the possibility of coincidental infection unrelated to the presentation. We additionally acknowledge the dramatic improvement after plasma could have been purely incidental or a result of multiple treatments (i.e., high-dose steroids and monthly IV Ig steroids) given during a short period of time. This case highlights the complex interaction between the brain, the immune system, and the SARS-CoV-2 virus, with sometimes paradoxical outcomes, echoed in the literature.^7,14 Further studies are required to explain the pathophysiology of SARS-CoV-2 infection, a disease fueled by host inflammatory response, ⁸ in immunocompromised patients.

Conclusion

Our case is a 54-year-old woman with isolated neurological manifestations associated with SARS-CoV-2 infection with unique radiological findings in the setting of CVID, with subsequent improvement following steroids, IV Ig, and CP. This case suggests that CP could be a safe and potentially beneficial adjuvant therapy for COVID-19 in patients with immunodeficiency. Further studies are required to elucidate the complex interaction between the brain, the immune system, and the SARS-CoV-2 virus.

ORCID iD

Hernan Nicolas Lemus https://orcid.org/0000-0003-2485-461X