Volume 211 - Issue 11

Extended detection and isolation of Murray Valley encephalitis virus in whole blood and urine

Authors:  Leon Caly, Natalie Davidson, Ram Ghimire, Brinthan Rajaratnam, Jonathan Marrow, Rob Baird, Ella M Meumann and Julian Druce

Med J Aust 2019; 211 (11): 499-500.e1. || doi: 10.5694/mja2.50423
Published online: 9 December 2019

A 23-year-old Indigenous woman from Arnhem Land, Northern Territory, presented with a 2- day history of fever, arthralgia and headache

Clinical record

A 23‐year‐old Indigenous woman from Arnhem Land, Northern Territory, presented with a 2‐day history of fever, arthralgia and headache. She had a background of systemic lupus erythematosus treated with daily azathioprine 100 mg and prednisolone 20 mg, and rheumatic heart disease treated with monthly intramuscular benzathine penicillin. After admission, she developed generalised seizures and confusion requiring intensive care unit admission and intubation. Magnetic resonance imaging (MRI) of the brain on Day 6 after admission demonstrated increased T2 and fluid‐attenuated inversion recovery (FLAIR) signal in the medial temporal lobes bilaterally (Box 1). Electroencephalogram was suggestive of episodic bifrontal delta wave activity not corresponding to seizure activity and generalised slowing. Cerebrospinal fluid (CSF) white cell count was 287 × 106/L with 90% mononuclear cells, CSF protein elevated at 1.49 g/L (normal range [NR], 0.15–0.45 g/L), and CSF glucose 4.5 mmol/L (NR, 2.7–4.2 mmol/L). The most likely diagnosis was thought to be encephalitis associated with systemic lupus erythematosus, and she initially received intravenous immunoglobulin and intravenous methylprednisolone 1 g daily. However, infection, including Murray Valley encephalitis virus (MVEV) — peak risk February to July — was considered in the differential diagnosis.

MVEV serology by haemagglutination inhibition revealed seroconversion based on two serum samples taken a month apart (< 1/10 to 1/640). MVEV IgM indirect immunofluorescence on CSF collected on Day 6 of admission was positive,1,2 but pan‐flavivirus and MVEV‐specific polymerase chain reaction (PCR) tests3 on the same CSF sample were negative. Pan‐flavivirus and MVEV‐specific PCR assays were positive on urine and whole blood collected on Day 6 after admission, but interestingly, plasma derived from this whole blood was negative (Box 2).

The patient had ongoing seizures that were treated with midazolam, levetiracetam and phenytoin, and remained intubated for 7 days. She subsequently had complete neurological recovery and was discharged home on a weaning course of prednisolone after one month in hospital, with no residual neurological deficits.

Notably, MVEV PCR was persistently positive in whole blood for 11 weeks and in urine for 5 weeks after presentation; after this, urine detection was sporadic. MVEV was only identified in ten of the 27 whole blood derived plasma specimens tested, with no positivity identified 6 weeks from initial diagnosis.

Aliquots of the Day 9 MVEV positive urine were inoculated onto cultured Vero cells to determine viral viability. Cytopathic effects were noted 5 days after inoculation, with culture supernatant found to have a lower cycle threshold (Ct) value (Ct = 15) by MVEV PCR compared with starting inoculum (Ct = 28), suggesting viral growth. Whole genome sequencing of purified viral RNA was performed on a MiSeq system (Illumina), with phylogenetic analysis of the prM‐E gene localising the isolate (MVE_2018) to the genotype 1, subtype A (G1A) clade (Box 3); which, to our knowledge, is the first circulating G1A detected in the NT, although few NT strains have previously been typed.4

Discussion

MVEV is a flavivirus transmitted by the Culex mosquito predominantly found in northern Western Australia and tropical NT, where it is maintained in a cycle of transmission between water birds and mosquitoes, with sporadic involvement of accidental hosts such as humans.1 Four MVEV genotypes have been recognised within Australia, with genotype 1 being currently the most dominant, although a single genotype 2 case was described in the NT in 2015.3

Infection with MVEV is most commonly asymptomatic or associated with mild febrile illness, but between 1/150 and 1/1000 people develop encephalitis. The mortality rate for encephalitis is 5–30%, with permanent neurological deficits occurring in up to 50% of survivors.5 Diagnosis is usually made with serology, including detection of IgM in CSF or a fourfold rise in serum antibody titres. Viral detection in CSF is possible early in the illness; however, it may be negative by the time neurological symptoms develop. Supportive radiological findings include bilateral grey matter hyperintensity on T2 imaging, usually in the thalamus and infrequently involving the temporal lobes. CSF typically shows elevated protein and leucocytosis, with a predominance of mononuclear cells.5 The case we present here was unusual in that the MRI findings were subtle and the neurological outcome was excellent, although longer term mild sequelae are sometimes identified.6,7

In this case, detection of MVEV in whole blood and urine was superior to plasma or CSF, suggesting these less invasive samples may be best for MVEV detection. Urine and whole blood have been successfully used for the PCR detection of other flaviviruses, including Zika virus, dengue virus and Japanese encephalitis virus.8,9 Our parallel testing of whole blood and plasma samples indicated that viral titres were significantly reduced in plasma compared with whole blood, implying that MVEV may associate with blood cells. It is our recommendation that when MVEV infection is suspected, unfractionated whole blood and urine be referred for MVEV molecular testing in addition to CSF and serum. We note that the prolonged period of MVEV detection in this case may be due to the patient's immunosuppressed state.

Lessons from practice

  • In the case of Murray Valley encephalitis virus (MVEV) infection, the virus can be detected by polymerase chain reaction (PCR) in whole blood for 11 weeks and in urine for 5 weeks. MVEV PCR on cerebrospinal fluid may be negative and only intermittently positive on plasma.
  • Less invasive sample types, such as whole blood and urine, may be optimal for the extended molecular detection of MVEV.
  • Molecular testing for MVEV should include whole blood and urine — it is important that whole blood is not fractionated.

Box 1 – Magnetic resonance imaging brain scan demonstrating increased T2 and fluid‐attenuated inversion recovery (FLAIR) signal in the medial temporal lobes, particularly on the right (arrow)


 

Box 2 – Polymerase chain reaction (PCR) detection of Murray Valley encephalitis virus (MVEV) in whole blood, urine and plasma over time. Timeline of events and real‐time PCR cycle threshold (Ct) values for whole blood, plasma, urine and cerebrospinal fluid


CSF = cerebrospinal fluid; LOD = limit of detection. * Denotes date of hospital admission (29 March). The LOD was a Ct value of 45 (black horizontal line). Increase in Ct denotes decrease in virus titre.

Box 3 – Neighbour‐joined tree of Murray Valley encephalitis virus (MVEV) prM‐E sequences confirms the MVE_2018 as genotype 1A (G1A)


AUS = Australia; PNG = Papua New Guinea. Viral genotype G1 subtypes 1A (orange) and 1B (blue), G2 (red) and G3/4 (purple) are indicated. G3/4 are no longer thought to be circulating within Australia. MVE_2018 may be the first circulating G1A detected domestically, outside of Western Australia. Scale bar indicates 0.05 nucleotide substitutions per site.


Authors


Competing interests


References


Provenance: Commissioned; externally peer reviewed.