Abstract
1846
Objectives Autoimmune encephalitis is characterized by brain parenchymal inflammation associated with the presence of auto-antibodies to neuronal cell surface proteins. Clinical presentation can vary widely, and distinguishing autoimmune encephalitis from other neurologic pathologies can present a significant clinical challenge. Early diagnosis is crucial to ensure that proper therapies are employed given the specific treatment strategies for managing autoimmune encephalitis. Imaging and laboratory testing play a key role in diagnosis, though results of antibody testing may not be available for several days or weeks. Imaging may lead to improved stratification and earlier implementation of appropriate treatment.
Methods Resting 18F-FDG PET-CT and MRI scans of the brain were performed in patients with suspected autoimmune encephalitis over an 8 year period. Imaging studies were separated by less than two weeks. Patients with antibody positive CSF assay were selected for analysis. Z-score maps of PET-CT scans were made using three-dimensional stereotactic surface projections with comparison to data from age group matched control subjects (Cortex ID, GE Healthcare). Brain metabolism was measured from filtered-back projection data normalized to the pons. MRI was performed by clinical protocol including standard T1, T2, FLAIR, and diffusion weighted sequences.
Results Twenty-one patients with antibody positive autoimmune encephalitis were analysed. PET-CT was abnormal in 90% of patients (19 of 21), and MRI was abnormal in only 33% of patients (7 of 21). Furthermore, 12 of the 14 patients (86%) with normal MRI findings were characterized as abnormal by PET-CT. PET-CT was categorized as normal or abnormal using quantitative Z-score mapping (abnormal > 2.0 standard deviations from age group matched controls). MRI was categorized as normal or abnormal by the presence or absence of T2/FLAIR, diffusion weighted, or enhancement abnormalities suggestive of encephalitis. Antibody assay results included positive tests for anti-voltage gated potassium channel (5), anti-N-methyl D-aspartate receptor (5), anti-glutamic acid decarboxylase (3), anti-Hu (3), anti-Ma (2), anti-leucine-rich glioma inactivated-1 (2), and anti-alpha-3 acetylcholine receptor (1) antibodies.
Conclusions In patients with antibody confirmed autoimmune encephalitis, FDG PET-CT is more often abnormal than MRI. PET-CT may serve as a more sensitive first line imaging test than MRI in patients for whom clinical suspicion is high. Additionally, patients with suspected autoimmune encephalitis and normal MRI findings may benefit from further imaging with FDG PET-CT.