Psychiatric Manifestation of Anti-LGI1 Encephalitis

Anti-leucine-rich glioma-inactivated 1 (LGI1) encephalitis is typically characterized by limbic encephalitis, faciobrachial dystonic seizures and hyponatremia. The frequency with which milder forms of anti-LGI1 encephalitis mimic isolated psychiatric syndromes, such as psychoses, or may lead to dementia if untreated, is largely unknown.

Here, the authors present a 50-year-old patient who had suffered from neurocognitive deficits and predominant delusions for over one and a half years. He reported a pronounced feeling of thirst, although he was drinking 10-20 liters of water each day, and he was absolutely convinced that he would die of thirst. Due to insomnia in the last five years, the patient took Z-drugs; later, he also abused alcohol. Two years prior to admission, he developed a status epilepticus which had been interpreted as a withdrawal seizure. In his serum, anti-LGI1 antibodies were repeatedly detected by different independent laboratories. Cerebrospinal fluid analyses revealed slightly increased white blood cell counts and evidence for blood-brain-barrier dysfunction. Magnetic resonance imaging showed hyperintensities mesio-temporally and in the right amygdala. In addition, there was a slight grey-white matter blurring. A cerebral [¹⁸F] fluorodeoxyglucose positron emission tomography (FDG-PET) examination of his brain showed moderate hypometabolism of the bilateral rostral mesial to medial frontal cortices. Treatment attempts with various psychotropic drugs remained unsuccessful in terms of symptom relief. After the diagnosis of probable chronified anti-LGI1 encephalitis was made, two glucocorticoid pulse treatments were performed, which led to a slight improvement of mood and neurocognitive deficits. Further therapy was not desired by the patient and his legally authorized parents.

This case study describes a patient with anti-LGI1 encephalitis in the chronified stage and a predominant long-lasting psychiatric course with atypical symptoms of psychosis and typical neurocognitive deficits. The patient's poor response to anti-inflammatory drugs was probably due to the delayed start of treatment. This delay in diagnosis and treatment may also have led to the FDG-PET findings, which were compatible with frontotemporal dementia ("state of damage"). In similar future cases, newly occurring epileptic seizures associated with psychiatric symptoms should trigger investigations for possible autoimmune encephalitis, even in patients with addiction or other pre-existing psychiatric conditions. This should in turn result in rapid organic clarification and-in positive cases-to anti-inflammatory treatment. Early treatment of anti-LGI1 encephalitis during the "inflammatory activity state" is crucial for overall prognosis and may avoid the development of dementia in some cases. Based on this case, the authors advocate the concept-long established in many chronic inflammatory diseases in rheumatology-of distinguishing between an "acute inflammatory state" and a "state of organ damage" in autoimmune psychosis resembling neurodegenerative mechanisms.