MRI findings in glutamic acid decarboxylase associated autoimmune epilepsy

Abstract

Purpose: Glutamic acid decarboxylase (GAD65) has been implicated in a number of autoimmune-associated neurologic syndromes, including autoimmune epilepsy. This study categorizes the spectrum of MRI findings in patients with a clinical diagnosis of autoimmune epilepsy and elevated serum GAD65 autoantibodies.

Methods: An institutional database search identified patients with elevated serum GAD65 antibodies and a clinical diagnosis of autoimmune epilepsy who had undergone brain MRI. Imaging studies were reviewed by three board-certified neuroradiologists and one neuroradiology fellow. Studies were evaluated for cortical/subcortical and hippocampal signal abnormality, cerebellar and cerebral volume loss, mesial temporal sclerosis, and parenchymal/leptomeningeal enhancement. The electronic medical record was reviewed for relevant clinical information and laboratory markers.

Results: A study cohort of 19 patients was identified. The majority of patients were female (84%), with a mean age of onset of 27 years. Serum GAD65 titers ranged from 33 to 4415 nmol/L (normal < 0.02 nmol/L). The most common presentation was medically intractable, complex partial seizures with temporal lobe onset. Parenchymal atrophy was the most common imaging finding (47%), with a subset of patients demonstrating cortical/subcortical parenchymal T2 hyperintensity (37%) or abnormal hippocampal signal (26%). No patients demonstrated abnormal parenchymal/leptomeningeal enhancement.

Conclusion: The most common MRI finding in GAD65-associated autoimmune epilepsy is disproportionate parenchymal atrophy for age, often associated with abnormal cortical/subcortical T2 hyperintensities. Hippocampal abnormalities are seen in a minority of patients. This constellation of findings in a patient with medically intractable epilepsy should raise the possibility of GAD65 autoimmunity.

Keywords: Autoimmune encephalopathy; Autoimmune epilepsy; GAD65; Glutamic acid decarboxylase.