Discussion

The differential diagnosis of acute cerebellar ataxia in a young adult includes vascular, inflammatory, toxic, neoplastic, paraneoplastic, genetic and infective causes (table 2). In this case, an adolescent developing tonsillitis shortly after the onset of ataxia raised the likelihood that EBV was the causative agent; this was subsequently confirmed on serological testing.

Acute cerebellar ataxia is a rare presentation of EBV infection and can occur without the systemic manifestations of EBV.1 In a review of acute cerebellar ataxia due to EBV, cerebellar ataxia was the sole manifestation in five out of the 16 patients.1 The age of onset ranged from 4 to 26 years (with the exception of a 52-year-old man) and there was a male predominance (12 of 16 cases). Patients had either a mild CSF lymphocytic pleocytosis (up to 44 cells/mm3) or no white blood cells in the CSF. All patients who were followed up had recovered within 3–13 weeks.1

It is not clear if the neurological manifestations of EBV result from direct tissue invasion or from a postinfectious autoimmune phenomenon. In our case, the neurological symptoms preceded the systemic features of EBV infection by several days, suggesting that they resulted from the acute infection rather than being a postinfectious phenomenon. EBV and its antibodies are often not found in the CSF (as in our case): in one case, however, EBV was detected directly in CSF cells.1,–,4

Although most cases of EBV cerebellitis have a benign outcome, there are several reports of life-threatening cerebellar swelling with acute cerebellitis from EBV. In these cases, an MR brain scan may show contrast enhancement or increased diffusion weighted signal in the cerebellar hemispheres, with subsequent fourth ventricle obstruction, hydrocephalus and transforaminal herniation.1 ,5 ,6 Our patient was alert and had a normal cerebellum on imaging. However, MRI may identify patients at risk of developing acute hydrocephalus and allow early intervention to prevent herniation. The number of patients treated with corticosteroids is too small to know whether they help. In the review of Erzurum et al, all patients followed up had recovered within 3–13 weeks; corticosteroids had been given in 4 out of 16 cases, without a clear benefit.1 Corticosteroids also did not prevent one patient with severe cerebellar swelling from deteriorating.5 Where cerebellar signs develop weeks after the onset of the EBV infection, and therefore with probable postinfectious cerebellitis, plasma exchange may help.7

In summary, we present an adolescent with acute cerebellar ataxia, which preceded the systemic manifestations of EBV infection. Neurological complications of EBV are rare, but also include Guillain–Barré syndrome, meningitis, encephalitis, cranial nerve palsies, optic neuritis and transverse myelitis.8 EBV can cause cerebellitis even without its systemic manifestations. The diagnosis is based on finding EBV VCA IgM antibodies in serum rather than CSF. In most cases, the prognosis is good, with complete recovery; however, clinicians must be aware of the risk of cerebellar swelling and hydrocephalus.