Enterovirus D68 has been associated with a poliomyelitis-like illness, notably during an outbreak in 2014, and particularly affecting children in the USA. We report a case of acute segmental flaccid paralysis with respiratory involvement in an adult in the UK, with enterovirus D68 detected in a sputum sample. MR imaging of cervical spinal cord showed a longitudinally extensive T2 hyperintensity in the anterior cord. Cerebrospinal fluid showed an elevated white cell count, predominantly lymphocytic, with otherwise normal constituents and negative viral PCRs. His respiratory function improved after intravenous immunoglobulin, suggesting that this may be useful in such cases. Clinicians should consider enterovirus D68 infection in the differential diagnosis of Guillain-Barré syndrome, particularly the pharyngeal–cervical–brachial variant.
- CLINICAL NEUROLOGY
- GUILLAIN-BARRE SYNDROME
- INFECTIOUS DISEASES
Statistics from Altmetric.com
A 45-year-old British Caucasian man gave a 5-day history of progressive left arm weakness, with neck and shoulders discomfort and vomiting for 3 days. On admission, he developed shortness of breath, difficulty swallowing, dry mouth and double vision. He had no sensory symptoms, sphincter disturbance or leg weakness. There had been a mild upper respiratory tract infection 2 weeks previously, with a violent cough and febrile symptoms a few days before the onset of weakness. He had not travelled outside the UK in recent months. He was usually fit and well and was an active long distance cyclist.
On examination, his temperature was 37.5°C and blood pressure 145/78 mm Hg. His respiratory rate was 20/min, and oxygen saturation was 94% on air. He had no meningism. There were bilateral partial sixth nerve palsies. He had markedly asymmetrical weakness of shoulder abduction (MRC grade 2 on the left and 4+ on the right), with grade 4 weakness of right elbow extension. His reflexes were variably brisk or absent in the upper limbs on repeated examinations, but normal in the lower limbs with downgoing plantars. There were no sensory signs. He became breathless on lying flat; forced vital capacity was 2.66 L (59% predicted).
Arterial blood gases showed a PaO2 of 8.84 kPa, PaCO2 of 6.22 kPa and pH of 7.407. He was monitored on intensive care overnight and the following day required intubation and ventilation due to type II respiratory failure.
CT scan of head was normal. His cerebrospinal fluid (CSF) white cell count was elevated (28 /µL; ≤5), but laboratory investigations were otherwise normal (table 1). His initial treatment was with acyclovir, ceftriaxone and clarithromycin and 3 days of intravenous methylprednisolone, 1 g each day, for a postinfectious or parainfectious inflammatory process such as acute disseminated encephalomyelitis.
MR scan of brain was normal. MR of cervical spine (3 days after presentation) showed subtle T2 hyperintensity in the anterior cord between the levels of C2 and C7, with abnormal signal in the anterior horn cells visible on cross sectional imaging at this level (figure 1). One week into the illness, routine sensory and motor nerve conduction were within normal limits; however, F-waves were infrequent from the left median and ulnar, suggesting a more proximal motor pathology (see online supplementary tables 1 and 2). Electromyography (EMG) found a loss of functional motor units from the weak left upper limb muscles but without positive evidence, in terms of spontaneous activity, for acute motor axon or anterior horn cell loss (see online supplementary table 3).
These findings were in keeping with the clinical suspicion of polyradiculopathy, but neurophysiological studies in early disease cannot distinguish motor axonal/anterior horn cell loss from proximal conduction block. We treated him for the treatable possibility with 5 days of intravenous immunoglobulin at 2 g/kg, for a working diagnosis of pharyngeal–cervical–brachial variant of Guillain-Barré Syndrome. His respiratory function improved, and he was extubated after 48 hours then gradually weaned off non-invasive ventilation; he was discharged home 11 days after first presentation.
Six weeks later, he reported a head drop that had since improved. He had developed striking and profound wasting of his neck muscles, particularly of his posterior shoulder girdle (figure 2). There was bilateral, asymmetrical scapular winging, weakness of neck extension and flexion and of shoulder abduction, internal and external rotations. He had no persistent respiratory disturbance, and he could lie flat without breathing difficulty, although his forced vital capacity remained reduced at 3.07 L (66% of predicted).
Repeat upper limb sensory and motor nerve conduction was unchanged but F-waves from the median and ulnar became normal. EMG now showed severe active and chronic neurogenic change involving the left upper cervical paraspinal muscles, trapezius and most of the left upper limb/shoulder girdle C5–C7 muscles sampled. There was minor chronic neurogenic change in the right triceps with subtle changes in deltoid and infraspinatus. These findings were now much more consistent with a localised and asymmetrical proximal motor axonal loss at the root or anterior horn cell level (see online supplementary tables 1, 2 and 3).
Further investigations available at this point included Lyme serology and antiganglioside antibodies to GM1 and GQ1b, which were negative (table 1); GT1a testing was not available. Respiratory virus serology was negative for adenovirus, influenza A and B, parainfluenza, respiratory syncytial virus, mycoplasma, Chlamydia spp. and Coxiella burnetii. However, respiratory PCRs on sputum detected rhinovirus and enterovirus D68. Enterovirus type D68 was confirmed by partial sequencing of the region encoding the capsular protein VP1.
Six months later, his clinical state was unchanged, leaving him with a significant weakness (figure 2). Follow-up MR imaging of the spinal cord was normal with no atrophy.
This patient developed a subacute segmental flaccid paralysis in the context of a respiratory tract infection with enterovirus D68, with an outcome reminiscent of poliomyelitis. We present the case as it unfolded; we considered several differentials including acute disseminated encephalomyelitis, lymphoma, brachial neuritis and the pharyngeal–cervical–brachial variant of Guillain-Barré syndrome (PCB-GBS).
PCB-GBS is a rare subtype of Guillain-Barré syndrome first described by Ropper in 19861 and later reviewed by Wakerly and Yuki.2 Patients usually present following an infection, with fairly symmetrical oropharyngeal and arm weakness, with or without sensory involvement, and commonly have IgG anti-GT1a or anti-GQ1b antibodies. Neurophysiology is variable and may show axonal features only. Our case, while sharing some features of PCB-GBS, is more in keeping with an acute anterior horn cell disease affecting predominantly the cervical myotomes. Features against PCB-GBS include the marked asymmetry, profound focal muscle wasting indicating severe denervation rather than reversible conduction block and absence of marked oropharyngeal weakness. The cellular CSF with normal protein further supports an acute infection rather than PCB-GBS, given that fewer than 10% of Guillain-Barré syndrome develop cell counts above 10 cells/µL.3 We also considered a brachial neuritis variant, where neurophysiology may show patchy pure motor denervation without sensory nerve changes, but we felt that the distribution and clinical features were outside of this spectrum.4 Pain was not a prominent feature, but he did report discomfort in his neck and shoulders that may have represented myalgic pain; there was no nuchal rigidity.
The first cases of a poliomyelitis-like neurological illness associated with enterovirus D68 were reported in North America and Canada, following a nationwide outbreak of more than 1000 cases of enterovirus D68 in 2014. This outbreak was associated with severe respiratory illness. Of twelve children in Colorado with this neurological illness, ten had flaccid limb weakness (asymmetrical in seven), six had bulbar involvement, five had cranial nerve involvement (sixth or seventh nerve) and ten had CSF pleocytosis.5 Ten children had MR imaging abnormalities, comprising confluent, longitudinally extensive spinal cord lesions of the central grey matter with predominant anterior horn cell involvement. Five of eleven nasopharyngeal specimens were positive for enterovirus D68, but it was not present in CSF, blood or rectal swabs. All 10 children with limb involvement were left with a persistent disability.
There were also three cases of acute flaccid paralysis in association with enterovirus D68 in 2014 in children in Europe (France and Norway).6 7 The two Norwegian cases were characterised by neck pain without nuchal rigidity. Enterovirus D68 was isolated from all three respiratory samples but not CSF—although CSF examination did show a pleocytosis of between 14 and 190 cells/µL. Known neurotrophic viruses such as poliovirus and enterovirus type 71 are often not isolated from CSF, with higher diagnostic yields coming from upper respiratory tract or gastrointestinal tract specimens.8 Thus, being unable to find enterovirus in our patient’s CSF is typical.
According to the American Centers for Disease Control and Prevention, there were 120 confirmed cases of acute flaccid myelitis in 2014, 21 in 2015 and 32 up to July 2016. (There are an estimated 10–15 million symptomatic cases of enterovirus infection (of various subtypes) annually in the USA.) The definition required MR imaging evidence of a spinal cord lesion in the grey matter, and patients had to be aged less than 21 years. Eighty per cent had CSF pleocytosis and many also had elevated CSF protein concentrations.9 The spike in 2014 coincided with the enterovirus D68 outbreak, but causation has not been determined.
Enterovirus D68 and poliovirus are members of the picornavirus family—small single-stranded positive-sense RNA viruses.10 Enterovirus D68 has many similarities to rhinoviruses, including acid sensitivity and a preference for lower temperatures (33°C rather than 37°C) for optimal replication, explaining why it is associated with respiratory rather than gastrointestinal presentations. It can be transmitted through contaminated surfaces or aerosolised particles. Enterovirus D68 infections show a seasonal variation, being most likely in early autumn and winter in temperate climates. Worldwide cases predominate in children, at median age of 4–5 years. The prevalence in adults is unclear, mainly because most studies have focused on children. Also, respiratory symptoms may be milder in adults and therefore may go unreported.
There is no specific treatment, nor any available vaccine, for enterovirus D68 infection. The infections are usually mild and self-limited. The frequency of neurological manifestations remains uncertain, particularly in adults. We are not aware of previously reported cases of acute, polio-like flaccid paralysis in adults in association with enterovirus D68. We suggest that adult neurologists should be aware of this rare cause of acute flaccid paralysis and consider sending sputum samples and/or viral throat swabs for respiratory PCR for enterovirus D68, particularly in cases suggesting the pharyngeal–cervical–brachial variant of Guillain-Barré syndrome, and possibly in the differential of brachial neuritis. It is reasonable to consider intravenous immunoglobulin treatment. In this case, we did not measure serum immunoglobulins before treatment; however, hypogammaglobulinaemia may predispose to enteroviral infection and, if present, would strengthen the rationale to give intravenous immunoglobulin.
Sputum samples and viral throat swabs (easily overlooked) can help with acute neurological presentations; their yield is highest early in the illness.
Enterovirus D68 is associated with an acute poliomyelitis-like flaccid paralysis with respiratory involvement in children but should also be considered in adults.
Consider intravenous immunoglobulin as treatment for enterovirus D68-associated flaccid paralysis.
Acknowledgements We thank Dr Justin Cross, consultant neuroradiologist, for his review of the MRI.
Contributors SRLS is the lead writer of the report and is the consultant neurologist responsible for the initial care of the patient. AM undertook the neurophysiological investigations, commented on the report and prepared Supplementary tables. DB contributed to writing the report, took over follow-up of the patient and provided the images.
Competing interests None declared.
Patient consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed. This paper was reviewed by Nick Davies, London, UK.
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.