First-episode psychosis with bilateral medial temporal lobe hyperintensities

To the Editor

A 23-year-old female university student presented to the hospital emergency department after waking screaming, hearing worsening ‘voices’. Assessment revealed third-person auditory hallucinations, religious delusions and catatonic posturing. She reported a 2-month period of greater religious involvement and 1 month of auditory hallucinations. There was no significant past psychiatric, medical, drug or alcohol history, or regular medications. There was a family history of her father having a period of heightened religiosity at a similar age. On examination, she was afebrile, with normal vital signs and neurological examination. However, possible lingual dystonia (tongue protrusion) was noted in the emergency department.

Blood tests including full blood count (FBC), electrolytes, urea and creatinine (EUC), calcium, magnesium, and phosphate (CMP), liver function tests (LFTs), thyroid-stimulating hormone (TSH), C-reactive protein (CRP) and beta-human chorionic gonadotropin (bHCG) were normal; anti-nuclear antibody (ANA) was 1:80, with negative extractable nuclear antigens (ENA)/double-stranded DNA (dsDNA)/anti-neutrophil cytoplasmic antibodies (ANCA). Creatinine kinase (CK) was elevated (1309) likely due to catatonic posturing and motor agitation. Computed tomography (CT) of the brain and electroencephalogram (EEG) were normal. Magnetic resonance imaging (MRI) of the brain demonstrated bilateral medial temporal lobe fluid attenuated inversion recovery (FLAIR) hyperintensities (Figure 1), suggesting the possibility of limbic encephalitis. The patient’s presentation, however, argued against this. A working diagnosis of first-episode psychosis (FEP) was made, with differentials of limbic and herpes simplex virus (HSV) encephalitis. Risperidone 0.5 mg daily was commenced, with resolution of symptoms over 1 week. Lumbar puncture was normal (including HSV polymerase chain reaction [PCR] negative). Pelvic ultrasound and whole-body positron emission tomography (PET) (F18-FDG) revealed no evidence of malignancy. Subsequent serum and cerebrospinal fluid (CSF) anti-neuronal Abs, anti-NMDAR Abs, anti-VGKC Abs and anti-VGCaC Abs were negative.

Figure 1.

Bilateral temporal lobe hyperintensities (white arrows).

Focal temporal lobe pathology is associated with secondary psychosis (Fricchione et al., 1995). However, routine MRI of FEP rarely demonstrates clinically relevant pathology, and bilateral temporal lobe hyperintensities are largely not described (Goulet et al., 2009). In our patient, this finding raised the possibility of an inflammatory aetiology of her psychopathology. Clinical diagnoses, however, must emphasise a thorough history with investigation results being interpreted against this information to maximise specificity and sensitivity.

These radiological findings may therefore be incidental, especially given the absence of other biological markers to support limbic or HSV encephalitis. We opted for antipsychotic pharmacotherapy, with recommendation for psychiatric and neurologic observation, with a plan to repeat neuroimaging if clinical deterioration occurs. This case raises the vexed issue of isolated neuroradiological abnormalities occurring in otherwise ‘typical’ psychiatric presentations and the need to determine a reasonable panel of investigations. Equally, it is important not to delay appropriate treatment.