Article Text

Download PDFPDF

Balancing immunosuppression and infection: recurrent enterovirus encephalitis in SLE
  1. Sanjay Cheema1,
  2. Eva Bunting2,
  3. Catriona Good3,
  4. Vijay Hajela4,
  5. Basil H Ridha1,
  6. Romi Anirban Saha1
  1. 1 Neurology, Brighton and Sussex University Hospitals NHS Trust, Haywards Heath, UK
  2. 2 Neurology, National Hospital for Neurology and Neurosurgery, London, UK
  3. 3 Neuroradiology, Brighton and Sussex University Hospitals NHS Trust, Brighton, UK
  4. 4 Rheumatology, Brighton and Sussex University Hospitals NHS Trust, Brighton, UK
  1. Correspondence to Dr Sanjay Cheema, Neurology, Brighton and Sussex University Hospitals NHS Trust, Haywards Heath, Sussex RH16 4EX, UK; sanjay.cheema{at}nhs.net

Abstract

A young woman with systemic lupus erythematosus (SLE) developed recurrent enterovirus meningoencephalitis while taking prednisolone, azathioprine and rituximab. After reducing the immunosuppression, she developed a central nervous system (CNS) flare of SLE, with enterovirus still present in the cerebrospinal fluid (CSF). There are no evidence-based specific treatments for enterovirus encephalitis, but she responded well to intravenous immunoglobulin alongside pulsed methylprednisolone and rituximab. This case highlights the difficulties in managing people with co-existing infective and autoimmune conditions, especially if each affects the CNS. A viral infection and SLE flare can resemble one another clinically, although here the radiological differentiation of CNS lupus versus enterovirus encephalitis helped to guide the diagnosis.

  • neurovirology
  • SLE
  • encephalitis
  • infectious diseases

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Systemic lupus erythematosus (SLE) is an immune-mediated disorder that can affect many organ systems, including the central nervous system (CNS). SLE directly impairs immune system function, and immunosuppressive therapy further increases vulnerability to infection. Having both SLE and infective disease together is often most severe when the CNS is affected.

Enterovirus is a rare cause of encephalitis with no specific treatment, usually occurring in immunosuppressed adults. There are several reports of enterovirus meningoencephalitis in the context of rituximab treatment. We describe a woman with SLE who developed recurrent enterovirus meningoencephalitis, followed by a CNS flare of SLE.

Case history

A 23-year-old woman was diagnosed with SLE in 2011 (age 16 years) after developing rash, alopecia, mouth ulcers, lymphopenia, thrombocytopenia and positive serum antinuclear antibodies. She was treated with prednisolone, azathioprine and hydroxychloroquine. In October 2015, she developed immune thrombocytopenic purpura: rituximab (375 mg/m2 in 2 doses 14 days apart) was added, with a further course in October 2016 after a cutaneous flare with pleuritis.

In July 2017, when working in a children’s nursery, she presented with a 3-day history of fever, vomiting, headache and neck pain. Neurological examination was normal and there was no clinical evidence of a systemic exacerbation of SLE. Her immunomodulatory medications at this time were azathioprine 150 mg and prednisolone 7.5 mg daily.

Full blood count, erythrocyte sedimentation rate, serum complement, and serum anti-double-stranded DNA and immunoglobulin G (IgG) concentrations were normal. Cerebrospinal fluid (CSF) examination showed a raised white blood cell count of 11 cells/µL (normal<5), raised protein of 0.60 g/L (normal 0.15–0.45) and low glucose at 2.4 mmol/L with corresponding plasma glucose of 4.9 mmol/L (normal CSF/plasma ratio>0.6). Viral PCR detected enterovirus RNA. MR scan of brain showed subtle corpus callosal signal change, thought to be caused by a transient metabolic process secondary to infective encephalitis (see figure 1A).

Figure 1

Series of MR scans of brain. The T2-weighted image in July 2017 (A) shows an area of signal change in the splenium of the corpus callosum thought to be a transient metabolic process secondary to viral encephalitis. In June 2018 (B), there was oedematous gyriform FLAIR (fluid-attenuated inversion recovery) signal change in both motor cortices and left cingulate gyrus, with restricted diffusion (C). This suggested that the changes were secondary to an acute lupus vasculitis, rather than enterovirus encephalitis.

We diagnosed enterovirus meningoencephalitis and managed her conservatively. We withheld azathioprine due to the infection and deranged liver function tests. She recovered well and repeat MR scan 1 week later was normal, with a resolution of the corpus callosal signal change. Two months following the infection, we changed her azathioprine to mycophenolate mofetil.

In February 2018, a few days after a viral upper respiratory tract infection, she developed confusion and photophobia, with no other focal neurological signs. Again, there was no evidence of a systemic SLE flare and blood markers were normal. MR scan of brain was normal. CSF examination showed a raised white cell count of 10 cells/µL, raised protein of 0.84 g/L and normal glucose of 2.0 mmol/L (plasma 3.7 mmol/L). Viral PCR again detected enterovirus RNA, shown on further testing to be the subtype enterovirus 71. We again managed her supportively and she fully recovered over a few days. We withheld the mycophenolate due to recurrent infections.

In June 2018, she presented with headache, vomiting, and speech and behavioural change developing over 48 hours and was treated empirically with intravenous ceftriaxone and aciclovir. MR scan of brain showed gyriform high signal within the perirolandic cortex bilaterally with restricted diffusion (figure 1B,C). CSF showed a raised white cell count of 17 cells/µL (80% lymphocytes), raised protein of 0.98 g/L and borderline low CSF glucose 2.5 mmol/L (plasma 4.8 mmol/L). Enterovirus subtype 71 was again detected on CSF viral PCR.

Two days later, her level of consciousness fell. She developed decorticate posturing and required intubation to protect her airway. Repeat MR scan of brain showed further increased signal change in the perirolandic cortex and left cingulate gyrus, with additional foci of restricted diffusion in the brainstem and superior and inferior surface of the cerebellum. There was no abnormal enhancement. MR angiography showed normal carotid arteries. However, there was subtle irregularity of the peripheral right occipital vessels, particularly the calcarine branch, which appeared attenuated; subtle irregularity of one of the left middle cerebral artery sylvian branches; and mild irregularity of the left distal M1 branch and right A1 branch.

The radiological appearances and clinical picture appeared more in keeping with CNS vasculitis related to SLE, rather than solely enterovirus meningoencephalitis. We, therefore, gave intravenous methylprednisolone 1 g per day for 3 days followed by 40 mg oral prednisolone.

In this case, we needed to treat CNS lupus at the time of intercurrent CNS enterovirus infection, and therefore risked worsening the infection with further immunosuppression. After discussion with our local virologist and immunologist, and gave intravenous immunoglobulin 2 g per kg body weight.

Repeat CSF examination 1 week later was negative for enterovirus. Repeat MR scan of brain at the same time showed that the areas of signal change were resolving, but with ischaemic damage in the left precentral gyrus and dorsal medulla.

Two weeks later, we gave a further course of rituximab (1 g infusion in 2 doses 14 days apart) and weaned the prednisolone by 5 mg per week to 5 mg daily. We plan a further dose of rituximab after 6 months. She has made a steady clinical recovery with no persistent neurological deficit.

Discussion

Enteroviruses comprise a diverse group of single-stranded RNA viruses, with over 100 human serotypes. Most enteroviruses cause mild coryzal symptoms but they can affect the CNS with significant morbidity and mortality. Enterovirus encephalitis is more common in children (especially those with agammaglobulinaemia) than adults, accounting for 1% of confirmed infective encephalitides in the UK. Enterovirus subtype 71 (as here) may cause epidemics of neurological disease in children.

There have been 16 reported cases of adult-onset enterovirus meningoencephalitis in the context of rituximab treatment, 14 of which were treated for haematological malignancies and 2 for autoimmune haematological disease.1 A total of 12 of the 16 cases reported had hypogammaglobulinaemia with a serum IgG concentration below 500 mg/dL.

There are no evidence-based proven medications to treat enteroviral infections. Pleconaril, designed to inhibit viral replication, is possibly effective in controlling common cold symptoms caused by viruses in the enterovirus and rhinovirus families. Pleconaril was effective in CNS enterovirus infections in small studies, but with subtypes other than enterovirus 71.

Some experimental findings suggest that intravenous immunoglobulin may help in enterovirus 71 encephalitis by modulating cytokine profiles,2 helping children with enterovirus CNS complications (encephalitis and acute disseminated encephalomyelitis); however, there is little controlled trial evidence supporting its routine use.

Of the 16 cases discussed above, 13 received intravenous immunoglobulin and 2 received pleconaril.1 Seven of the cases died, of which four had received immunoglobulin. The limited evidence makes it difficult to know if intravenous immunoglobulin is effective.

Viral pathogens have been implemented in the immune-mediated pathogenesis of SLE and act as triggers for some SLE flares. It is possible that the enterovirus infection triggered this patient’s CNS lupus flare. There was no flare during the first two episodes of meningoencephalitis while she was heavily immunosuppressed, but only the third episode when taking only hydroxychloroquine and low-dose prednisolone.

It is sometimes difficult to distinguish viral infection from a flare of SLE, as the symptoms, signs and blood investigations can be similar. In this case, neuroimaging helped to guide the diagnosis. The typical imaging features of enterovirus encephalitis include oedematous lesions within the brainstem, dentate nuclei and spinal cord, and less commonly within the thalami and basal ganglia, whereas restricted diffusion and contrast enhancement are uncommon. MR imaging in active CNS lupus typically shows periventricular white matter lesions and or cortical infarcts with restricted diffusion, and/or some haemorrhagic features on susceptibility weighted imaging. Active lesions often enhance. In this case, the imaging showed multiple areas of restricted diffusion at different time points, with associated vascular abnormalities but no brainstem or cerebellar lesions.

Although not conclusive evidence, the imaging features raise the possibility of a vasculitis associated with SLE. The major histopathological studies of SLE suggest that a true vasculitis can occur but is comparatively rare. The most common CNS changes are small vessel infarction in the cerebral cortex and brainstem due to a vasculopathy. Conversely, the vascular changes could represent a vasculitis triggered by the enterovirus infection, entirely separate from the SLE.

Patients suspected of having an acute CNS SLE flare should receive high-dose corticosteroids, followed by powerful immunotherapy, for example, either with cyclophosphamide or rituximab.

Key points

  • Enterovirus rarely causes encephalitis in the general population, but the risk is greatly increased in people taking immunomodulatory medications or those with hypogammaglobulinaemia.

  • Although there is no specific evidence for treatment, both intravenous immunoglobulin and pleconaril have been used for enterovirus encephalitis.

  • Infection can trigger flares in autoimmune diseases, such as systemic lupus erythematosus (SLE).

  • Vasculopathy causing small vessel infarction is the most common pathological process in central nervous system SLE.

  • Where the distinction is unclear between an inflammatory or infective process, certain radiological changes can suggest immune-mediated vasculopathy or vasculitis, rather than a primary infective cause.

References

Footnotes

  • Contributors All the authors were involved in the patient’s clinical care. SC and EB wrote the first draft and performed the literature review. All the authors were involved in editing the manuscript and approval of the final version submitted for publication.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed by Fady Joseph, Newport, UK.