Herpes simplex encephalitis mimicking a primary intracerebral haemorrhage

A previously healthy, right-handed 52-year-old woman developed a 2-day history of a constant generalised headache and confusion. She then had a generalised tonic-clonic seizure, which self-terminated after 2 minutes.

Her vital signs were normal. She was disoriented but lacked lateralising signs. Blood tests including glucose, complete blood count, renal function and inflammatory markers were normal.

Computed tomography (CT) of the brain showed acute left anterior temporal haematoma (Figure 1). The CT-angiogram showed no vascular abnormality and the venous phase showed completely patent dural venous sinuses.

She was admitted for observation and further investigations given the unclear aetiology of the bleed, and was commenced on levetiracetam 500mg b.d. She deteriorated further with increasing confusion and agitation. This was partly attributed to the intracerebral haemorrhage and the effect of the seizure.

Magnetic resonance imaging (MRI) of the brain performed on the second day revealed an asymmetric bitemporal T2 hyperintensity (affecting the left side more than the right), suggestive of herpes simplex virus (HSV) encephalitis (Figure 2). Subsequently, lumbar puncture (LP) confirmed the radiological suspicion with detection of a positive polymerase chain reaction (PCR) for type 1 HSV. There were 895 white cells (95% lymphocyte), elevated protein at 1.59g/l in the cerebrospinal fluid (CSF) and a normal CSF-to-serum glucose ratio.

She received intravenous acyclovir for 14 days and a repeat CSF following completion of treatment showed no evidence of active HSV infection with a negative PCR result. A reduction in the degree of pleocytosis (115 cells, 100% lymphocytes) and protein level (0.50g/l) was also noted.

She has had no further seizures. She has regained orientation and has demonstrated an improved ability to engage in a lengthier, more meaningful conversation. She scored 16/30 on the Montreal Cognitive Assessment with deficits noted in short-term memory, verbal fluency and attention reflecting the function of the structures involved.

There are limited reports of HSV encephalitis presenting with acute hemorrhage.^1,2 In contrast to our patient, all the reported cases demonstrated some clues for the presence of an underlying infection including a high temperature, raised peripheral white cell count or other inflammatory markers at presentation. Due to the delay in making an accurate diagnosis in these cases, prognosis is often unfavourable with increased morbidity and mortality.³

LP may be contraindicated in patients with a large haemorrhage due to risk of uncal herniation, and diagnosis may only be confirmed with a brain biopsy after decompressive hemicraniectomy or at post-mortem examination.^4,5

In our patient, the absence of infective symptoms, normal body temperature and blood test results, coupled with the acute anterior temporal haematoma seen on initial CT, diverted our thinking from considering an underlying infection.

Provided it is considered safe, we propose that LP could be considered in people presenting with cerebral haemorrhage of unclear aetiology when there is a clue for an infective aetiology. LP should also be considered in the acute phase when there is a clinico-radiological mismatch. Our patient had a small-size, early subacute haematoma at presentation (dark on T2, with a bright edge on T1), and in the absence of non-convulsive status, the imaging finding would not be expected to cause her worsening clinical picture without a hidden parenchymal involvement. Hence, in patients with unexplained cerebral haemorrhage, an MRI could be helpful by assessing the presence of any concealed disease and guiding further management. Periodic lateralised epileptiform discharges (PLEDs) are highly consistent with HSV in the appropriate clinical settings, and therefore an electroencephalogram can be extremely helpful in supporting the diagnosis of HSV encephalitis.⁶ However, intravenous acyclovir must be administered immediately for all suspected cases of HSV encephalitis without waiting for the results of imaging or CSF.

Clinicians should consider HSV encephalitis as a possible underlying aetiology for cerebral haemorrhage affecting the typical locations of HSV encephalitis, involving primarily the medial temporal lobes but also insular cortex, inferior frontal or cingulate gyrus. Swift recognition and initiation of appropriate antiviral therapy are crucial to reduce morbidity and mortality.

View Figure 1–2.